mirror of
https://github.com/JayDDee/cpuminer-opt.git
synced 2025-09-17 23:44:27 +00:00
Compare commits
19 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
d6e8d7a46e | ||
|
|
71d6b97ee8 | ||
|
|
b2331375a3 | ||
|
|
7fec680835 | ||
|
|
1b0a5aadf6 | ||
|
|
0a3c52810e | ||
|
|
4d4386a374 | ||
|
|
ce259b915a | ||
|
|
02202ab803 | ||
|
|
77c5ae80ab | ||
|
|
eb3f57bfc7 | ||
|
|
e1aead3c76 | ||
|
|
bfd1c002f9 | ||
|
|
9edc650042 | ||
|
|
218cef337a | ||
|
|
9ffce7bdb7 | ||
|
|
c7efa50aad | ||
|
|
dd5e552357 | ||
|
|
f449c6725f |
123
INSTALL_LINUX
Normal file
123
INSTALL_LINUX
Normal file
@@ -0,0 +1,123 @@
|
||||
|
||||
|
||||
Requirements:
|
||||
|
||||
Intel Core2 or newer, or AMD Steamroller or newer CPU. ARM CPUs are not
|
||||
supported.
|
||||
64 bit Linux operating system. Apple is not supported.
|
||||
|
||||
Building on linux prerequisites:
|
||||
|
||||
It is assumed users know how to install packages on their system and
|
||||
be able to compile standard source packages. This is basic Linux and
|
||||
beyond the scope of cpuminer-opt. Regardless compiling is trivial if you
|
||||
follow the instructions.
|
||||
|
||||
Make sure you have the basic development packages installed.
|
||||
Here is a good start:
|
||||
|
||||
http://askubuntu.com/questions/457526/how-to-install-cpuminer-in-ubuntu
|
||||
|
||||
Install any additional dependencies needed by cpuminer-opt. The list below
|
||||
are some of the ones that may not be in the default install and need to
|
||||
be installed manually. There may be others, read the error messages they
|
||||
will give a clue as to the missing package.
|
||||
|
||||
The following command should install everything you need on Debian based
|
||||
distributions such as Ubuntu:
|
||||
|
||||
sudo apt-get install build-essential libssl-dev libcurl4-openssl-dev libjansson-dev libgmp-dev automake zlib1g-dev
|
||||
|
||||
build-essential (Development Tools package group on Fedora)
|
||||
automake
|
||||
libjansson-dev
|
||||
libgmp-dev
|
||||
libcurl4-openssl-dev
|
||||
libssl-dev
|
||||
lib-thread
|
||||
zlib1g-dev
|
||||
|
||||
SHA support on AMD Ryzen CPUs requires gcc version 5 or higher and
|
||||
openssl 1.1.0e or higher. Add one of the following, depending on the
|
||||
compiler version, to CFLAGS:
|
||||
"-march=native" or "-march=znver1" or "-msha".
|
||||
|
||||
Additional instructions for static compilalation can be found here:
|
||||
https://lxadm.com/Static_compilation_of_cpuminer
|
||||
Static builds should only considered in a homogeneous HW and SW environment.
|
||||
Local builds will always have the best performance and compatibility.
|
||||
|
||||
Extract cpuminer source.
|
||||
|
||||
tar xvzf cpuminer-opt-x.y.z.tar.gz
|
||||
cd cpuminer-opt-x.y.z
|
||||
|
||||
Run ./build.sh to build on Linux or execute the following commands.
|
||||
|
||||
./autogen.sh
|
||||
CFLAGS="-O3 -march=native -Wall" ./configure --with-curl
|
||||
make
|
||||
|
||||
Start mining.
|
||||
|
||||
./cpuminer -a algo -o url -u username -p password
|
||||
|
||||
Windows
|
||||
|
||||
Precompiled Windows binaries are built on a Linux host using Mingw
|
||||
with a more recent compiler than the following Windows hosted procedure.
|
||||
|
||||
Building on Windows prerequisites:
|
||||
|
||||
msys
|
||||
mingw_w64
|
||||
Visual C++ redistributable 2008 X64
|
||||
openssl
|
||||
|
||||
Install msys and mingw_w64, only needed once.
|
||||
|
||||
Unpack msys into C:\msys or your preferred directory.
|
||||
|
||||
Install mingw_w64 from win-builds.
|
||||
Follow instructions, check "msys or cygwin" and "x86_64" and accept default
|
||||
existing msys instalation.
|
||||
|
||||
Open a msys shell by double clicking on msys.bat.
|
||||
Note that msys shell uses linux syntax for file specifications, "C:\" is
|
||||
mounted at "/c/".
|
||||
|
||||
Add mingw bin directory to PATH variable
|
||||
PATH="/c/msys/opt/windows_64/bin/:$PATH"
|
||||
|
||||
Instalation complete, compile cpuminer-opt.
|
||||
|
||||
Unpack cpuminer-opt source files using tar from msys shell, or using 7zip
|
||||
or similar Windows program.
|
||||
|
||||
In msys shell cd to miner directory.
|
||||
cd /c/path/to/cpuminer-opt
|
||||
|
||||
Run build.sh to build on Windows or execute the following commands.
|
||||
|
||||
./autogen.sh
|
||||
CFLAGS="-O3 -march=native -Wall" ./configure --with-curl
|
||||
make
|
||||
|
||||
Start mining
|
||||
|
||||
cpuminer.exe -a algo -o url -u user -p password
|
||||
|
||||
The following tips may be useful for older AMD CPUs.
|
||||
|
||||
AMD CPUs older than Steamroller, including Athlon x2 and Phenom II x4, are
|
||||
not supported by cpuminer-opt due to an incompatible implementation of SSE2
|
||||
on these CPUs. Some algos may crash the miner with an invalid instruction.
|
||||
Users are recommended to use an unoptimized miner such as cpuminer-multi.
|
||||
|
||||
Some users with AMD CPUs without AES_NI have reported problems compiling
|
||||
with build.sh or "-march=native". Problems have included compile errors
|
||||
and poor performance. These users are recommended to compile manually
|
||||
specifying "-march=btver1" on the configure command line.
|
||||
|
||||
Support for even older x86_64 without AES_NI or SSE2 is not availble.
|
||||
|
||||
173
INSTALL_WINDOWS
Normal file
173
INSTALL_WINDOWS
Normal file
@@ -0,0 +1,173 @@
|
||||
Instructions for compiling cpuminer-opt for Windows.
|
||||
|
||||
|
||||
Windows compilation using Visual Studio is not supported. Mingw64 is
|
||||
used on a Linux system (bare metal or virtual machine) to cross-compile
|
||||
cpuminer-opt executable binaries for Windows.
|
||||
|
||||
These instructions were written for Debian and Ubuntu compatible distributions
|
||||
but should work on other major distributions as well. However some of the
|
||||
package names or file paths may be different.
|
||||
|
||||
It is assumed a Linux system is already available and running. And the user
|
||||
has enough Linux knowledge to find and install packages and follow these
|
||||
instructions.
|
||||
|
||||
First it is a good idea to create new user specifically for cross compiling.
|
||||
It keeps all mingw stuff contained and isolated from the rest of the system.
|
||||
|
||||
Step by step...
|
||||
|
||||
1. Install necessary packages from the distribution's repositories.
|
||||
|
||||
Refer to Linux compile instructions and install required packages.
|
||||
|
||||
Additionally, install mingw-64.
|
||||
|
||||
sudo apt-get install mingw-w64
|
||||
|
||||
|
||||
2. Create a local library directory for packages to be compiled in the next
|
||||
step. Recommended location is $HOME/usr/lib/
|
||||
|
||||
|
||||
3. Download and build other packages for mingw that don't have a mingw64
|
||||
version available in the repositories.
|
||||
|
||||
Download the following source code packages from their respective and
|
||||
respected download locations, copy them to ~/usr/lib/ and uncompress them.
|
||||
|
||||
openssl
|
||||
curl
|
||||
gmp
|
||||
|
||||
In most cases the latest vesrion is ok but it's safest to download
|
||||
the same major and minor version as included in your distribution.
|
||||
|
||||
Run the following commands or follow the supplied instructions.
|
||||
Do not run "make install" unless you are using ~/usr/lib, which isn't
|
||||
recommended.
|
||||
|
||||
Some instructions insist on running "make check". If make check fails
|
||||
it may still work, YMMV.
|
||||
|
||||
You can speed up "make" by using all CPU cores available with "-j n" where
|
||||
n is the number of CPU threads you want to use.
|
||||
|
||||
openssl:
|
||||
|
||||
./Configure mingw64 shared --cross-compile-prefix=x86_64-w64-mingw32
|
||||
make
|
||||
|
||||
curl:
|
||||
|
||||
./configure --with-winssl --with-winidn --host=x86_64-w64-mingw32
|
||||
make
|
||||
|
||||
gmp:
|
||||
|
||||
./configure --host=x86_64-w64-mingw32
|
||||
make
|
||||
|
||||
|
||||
|
||||
4. Tweak the environment.
|
||||
|
||||
This step is required everytime you login or the commands can be added to
|
||||
.bashrc.
|
||||
|
||||
Define some local variables to point to local library.
|
||||
|
||||
export LOCAL_LIB="$HOME/usr/lib"
|
||||
|
||||
export LDFLAGS="-L$LOCAL_LIB/curl/lib/.libs -L$LOCAL_LIB/gmp/.libs -L$LOCAL_LIB/openssl"
|
||||
|
||||
export CONFIGURE_ARGS="--with-curl=$LOCAL_LIB/curl --with-crypto=$LOCAL_LIB/openssl --host=x86_64-w64-mingw32"
|
||||
|
||||
Create a release directory and copy some dll files previously built.
|
||||
This can be done outside of cpuminer-opt and only needs to be done once.
|
||||
If the release directory is in cpuminer-opt directory it needs to be
|
||||
recreated every a source package is decompressed.
|
||||
|
||||
mkdir release
|
||||
cp /usr/x86_64-w64-mingw32/lib/zlib1.dll release/
|
||||
cp /usr/x86_64-w64-mingw32/lib/libwinpthread-1.dll release/
|
||||
cp /usr/lib/gcc/x86_64-w64-mingw32/7.3-win32/libstdc++-6.dll release/
|
||||
cp /usr/lib/gcc/x86_64-w64-mingw32/7.3-win32/libgcc_s_seh-1.dll release/
|
||||
cp $LOCAL_LIB/openssl/libcrypto-1_1-x64.dll release/
|
||||
cp $LOCAL_LIB/curl/lib/.libs/libcurl-4.dll release/
|
||||
|
||||
|
||||
|
||||
The following steps need to be done every time a new source package is
|
||||
opened.
|
||||
|
||||
5. Download cpuminer-opt
|
||||
|
||||
Download the latest source code package of cpumuner-opt to your desired
|
||||
location. .zip or .tar.gz, your choice.
|
||||
|
||||
https://github.com/JayDDee/cpuminer-opt/releases
|
||||
|
||||
Decompress and change to the cpuminer-opt directory.
|
||||
|
||||
|
||||
|
||||
6. Prepare to compile
|
||||
|
||||
Create a link to the locally compiled version of gmp.h
|
||||
|
||||
ln -s $LOCAL_LIB/gmp-version/gmp.h ./gmp.h
|
||||
|
||||
Edit configure.ac to fix lipthread package name.
|
||||
|
||||
sed -i 's/"-lpthread"/"-lpthreadGC2"/g' configure.ac
|
||||
|
||||
|
||||
7. Compile
|
||||
|
||||
you can use the default compile if you intend to use cpuminer-opt on the
|
||||
same CPU and the virtual machine supports that architecture.
|
||||
|
||||
./build.sh
|
||||
|
||||
Otherwise you can compile manually while setting options in CFLAGS.
|
||||
|
||||
Some common options:
|
||||
|
||||
To compile for a specific CPU architecture:
|
||||
|
||||
CFLAGS="-O3 -march=znver1 -Wall" ./configure --with-curl
|
||||
|
||||
This will compile for AMD Ryzen.
|
||||
|
||||
You can compile more generically for a set of specific CPU features
|
||||
if you know what features you want:
|
||||
|
||||
CFLAGS="-O3 -maes -msse4.2 -Wall" ./configure --with-curl
|
||||
|
||||
This will compile for an older CPU that does not have AVX.
|
||||
|
||||
You can find several examples in build-allarch.sh
|
||||
|
||||
If you have a CPU with more than 64 threads and Windows 7 or higher you
|
||||
can enable the CPU Groups feature:
|
||||
|
||||
-D_WIN32_WINNT==0x0601
|
||||
|
||||
Once you have run configure successfully run make with n CPU threads:
|
||||
|
||||
make -j n
|
||||
|
||||
Copy cpuminer.exe to the release directory, compress and copy the release
|
||||
directory to a Windows system and run cpuminer.exe from the command line.
|
||||
|
||||
Run cpuminer
|
||||
|
||||
In a command windows change directories to the unzipped release folder.
|
||||
to get a list of all options:
|
||||
|
||||
cpuminer.exe --help
|
||||
|
||||
Command options are specific to where you mine. Refer to the pool's
|
||||
instructions on how to set them.
|
||||
63
Makefile.am
63
Makefile.am
@@ -31,14 +31,22 @@ cpuminer_SOURCES = \
|
||||
crypto/hash.c \
|
||||
crypto/aesb.c \
|
||||
crypto/magimath.cpp \
|
||||
algo/argon2/argon2a.c \
|
||||
algo/argon2/ar2/argon2.c \
|
||||
algo/argon2/ar2/opt.c \
|
||||
algo/argon2/ar2/cores.c \
|
||||
algo/argon2/ar2/ar2-scrypt-jane.c \
|
||||
algo/argon2/ar2/blake2b.c \
|
||||
algo/argon2/argon2a/argon2a.c \
|
||||
algo/argon2/argon2a/ar2/argon2.c \
|
||||
algo/argon2/argon2a/ar2/opt.c \
|
||||
algo/argon2/argon2a/ar2/cores.c \
|
||||
algo/argon2/argon2a/ar2/ar2-scrypt-jane.c \
|
||||
algo/argon2/argon2a/ar2/blake2b.c \
|
||||
algo/argon2/argon2d/argon2d-gate.c \
|
||||
algo/argon2/argon2d/blake2/blake2b.c \
|
||||
algo/argon2/argon2d/argon2d/argon2.c \
|
||||
algo/argon2/argon2d/argon2d/core.c \
|
||||
algo/argon2/argon2d/argon2d/opt.c \
|
||||
algo/argon2/argon2d/argon2d/argon2d_thread.c \
|
||||
algo/argon2/argon2d/argon2d/encoding.c \
|
||||
algo/blake/sph_blake.c \
|
||||
algo/blake/blake-hash-4way.c \
|
||||
algo/blake/blake256-hash-4way.c \
|
||||
algo/blake/blake512-hash-4way.c \
|
||||
algo/blake/blake-gate.c \
|
||||
algo/blake/blake.c \
|
||||
algo/blake/blake-4way.c \
|
||||
@@ -60,14 +68,15 @@ cpuminer_SOURCES = \
|
||||
algo/blake/pentablake-4way.c \
|
||||
algo/blake/pentablake.c \
|
||||
algo/bmw/sph_bmw.c \
|
||||
algo/bmw/bmw-hash-4way.c \
|
||||
algo/bmw/bmw256-hash-4way.c \
|
||||
algo/bmw/bmw512-hash-4way.c \
|
||||
algo/bmw/bmw256.c \
|
||||
algo/cryptonight/cryptolight.c \
|
||||
algo/cryptonight/cryptonight-common.c\
|
||||
algo/cryptonight/cryptonight-aesni.c\
|
||||
algo/cryptonight/cryptonight.c\
|
||||
algo/cubehash/sph_cubehash.c \
|
||||
algo/cubehash/sse2/cubehash_sse2.c\
|
||||
algo/cubehash/cubehash_sse2.c\
|
||||
algo/cubehash/cube-hash-2way.c \
|
||||
algo/echo/sph_echo.c \
|
||||
algo/echo/aes_ni/hash.c\
|
||||
@@ -109,26 +118,29 @@ cpuminer_SOURCES = \
|
||||
algo/luffa/luffa-hash-2way.c \
|
||||
algo/lyra2/lyra2.c \
|
||||
algo/lyra2/sponge.c \
|
||||
algo/lyra2/lyra2rev2-gate.c \
|
||||
algo/lyra2/lyra2-gate.c \
|
||||
algo/lyra2/lyra2rev2.c \
|
||||
algo/lyra2/lyra2rev2-4way.c \
|
||||
algo/lyra2/lyra2rev3.c \
|
||||
algo/lyra2/lyra2rev3-4way.c \
|
||||
algo/lyra2/lyra2re.c \
|
||||
algo/lyra2/lyra2z-gate.c \
|
||||
algo/lyra2/lyra2z.c \
|
||||
algo/lyra2/lyra2z-4way.c \
|
||||
algo/lyra2/lyra2z330.c \
|
||||
algo/lyra2/lyra2h-gate.c \
|
||||
algo/lyra2/lyra2h.c \
|
||||
algo/lyra2/lyra2h-4way.c \
|
||||
algo/lyra2/allium-gate.c \
|
||||
algo/lyra2/allium-4way.c \
|
||||
algo/lyra2/allium.c \
|
||||
algo/lyra2/phi2-4way.c \
|
||||
algo/lyra2/phi2.c \
|
||||
algo/m7m.c \
|
||||
algo/neoscrypt/neoscrypt.c \
|
||||
algo/nist5/nist5-gate.c \
|
||||
algo/nist5/nist5-4way.c \
|
||||
algo/nist5/nist5.c \
|
||||
algo/nist5/zr5.c \
|
||||
algo/panama/sph_panama.c \
|
||||
algo/radiogatun/sph_radiogatun.c \
|
||||
algo/pluck.c \
|
||||
algo/quark/quark-gate.c \
|
||||
algo/quark/quark.c \
|
||||
@@ -136,6 +148,9 @@ cpuminer_SOURCES = \
|
||||
algo/quark/anime-gate.c \
|
||||
algo/quark/anime.c \
|
||||
algo/quark/anime-4way.c \
|
||||
algo/quark/hmq1725-gate.c \
|
||||
algo/quark/hmq1725-4way.c \
|
||||
algo/quark/hmq1725.c \
|
||||
algo/qubit/qubit-gate.c \
|
||||
algo/qubit/qubit.c \
|
||||
algo/qubit/qubit-2way.c \
|
||||
@@ -152,12 +167,18 @@ cpuminer_SOURCES = \
|
||||
algo/sha/sph_sha2.c \
|
||||
algo/sha/sph_sha2big.c \
|
||||
algo/sha/sha2-hash-4way.c \
|
||||
algo/sha/sha256_hash_11way.c \
|
||||
algo/sha/sha2.c \
|
||||
algo/sha/sha256t-gate.c \
|
||||
algo/sha/sha256t-4way.c \
|
||||
algo/sha/sha256t.c \
|
||||
algo/sha/sha256q-4way.c \
|
||||
algo/sha/sha256q.c \
|
||||
algo/shabal/sph_shabal.c \
|
||||
algo/shabal/shabal-hash-4way.c \
|
||||
algo/shavite/sph_shavite.c \
|
||||
algo/shavite/sph-shavite-aesni.c \
|
||||
algo/shavite/shavite-hash-2way.c \
|
||||
algo/shavite/shavite.c \
|
||||
algo/simd/sph_simd.c \
|
||||
algo/simd/nist.c \
|
||||
@@ -231,19 +252,25 @@ cpuminer_SOURCES = \
|
||||
algo/x15/x15-gate.c \
|
||||
algo/x15/x15.c \
|
||||
algo/x15/x15-4way.c \
|
||||
algo/x16/x16r-gate.c \
|
||||
algo/x16/x16r.c \
|
||||
algo/x16/x16r-4way.c \
|
||||
algo/x17/x17-gate.c \
|
||||
algo/x17/x17.c \
|
||||
algo/x17/x17-4way.c \
|
||||
algo/x17/xevan-gate.c \
|
||||
algo/x17/xevan.c \
|
||||
algo/x17/xevan-4way.c \
|
||||
algo/x17/x16r-gate.c \
|
||||
algo/x17/x16r.c \
|
||||
algo/x17/x16r-4way.c \
|
||||
algo/x17/hmq1725.c \
|
||||
algo/x17/sonoa-gate.c \
|
||||
algo/x17/sonoa-4way.c \
|
||||
algo/x17/sonoa.c \
|
||||
algo/x20/x20r.c \
|
||||
algo/yescrypt/yescrypt.c \
|
||||
algo/yescrypt/sha256_Y.c \
|
||||
algo/yescrypt/yescrypt-best.c
|
||||
algo/yescrypt/yescrypt-best.c \
|
||||
algo/yespower/yespower.c \
|
||||
algo/yespower/sha256_p.c \
|
||||
algo/yespower/yespower-opt.c
|
||||
|
||||
disable_flags =
|
||||
|
||||
|
||||
156
README.md
156
README.md
@@ -7,11 +7,17 @@ All of the code is believed to be open and free. If anyone has a
|
||||
claim to any of it post your case in the cpuminer-opt Bitcoin Talk forum
|
||||
or by email.
|
||||
|
||||
Miner programs are often flagged as malware by antivirus programs. This is
|
||||
a false positive, they are flagged simply because they are cryptocurrency
|
||||
miners. The source code is open for anyone to inspect. If you don't trust
|
||||
the software, don't use it.
|
||||
|
||||
https://bitcointalk.org/index.php?topic=1326803.0
|
||||
|
||||
mailto://jayddee246@gmail.com
|
||||
|
||||
See file RELEASE_NOTES for change log and compile instructions.
|
||||
See file RELEASE_NOTES for change log and INSTALL_LINUX or INSTALL_WINDOWS
|
||||
for compile instructions.
|
||||
|
||||
Requirements
|
||||
------------
|
||||
@@ -40,83 +46,97 @@ MacOS, OSx and Android are not supported.
|
||||
Supported Algorithms
|
||||
--------------------
|
||||
|
||||
allium Garlicoin
|
||||
anime Animecoin
|
||||
argon2
|
||||
axiom Shabal-256 MemoHash
|
||||
allium Garlicoin
|
||||
anime Animecoin
|
||||
argon2 Argon2 coin (AR2)
|
||||
argon2d250 argon2d-crds, Credits (CRDS)
|
||||
argon2d500 argon2d-dyn, Dynamic (DYN)
|
||||
argon2d4096 argon2d-uis, Unitus, (UIS)
|
||||
axiom Shabal-256 MemoHash
|
||||
bastion
|
||||
blake Blake-256 (SFR)
|
||||
blakecoin blake256r8
|
||||
blake2s Blake-2 S
|
||||
bmw BMW 256
|
||||
c11 Chaincoin
|
||||
cryptolight Cryptonight-light
|
||||
cryptonight cryptonote, Monero (XMR)
|
||||
blake Blake-256 (SFR)
|
||||
blakecoin blake256r8
|
||||
blake2s Blake-2 S
|
||||
bmw BMW 256
|
||||
c11 Chaincoin
|
||||
cryptolight Cryptonight-light
|
||||
cryptonight
|
||||
cryptonightv7 Monero (XMR)
|
||||
decred
|
||||
deep Deepcoin (DCN)
|
||||
dmd-gr Diamond-Groestl
|
||||
drop Dropcoin
|
||||
fresh Fresh
|
||||
groestl Groestl coin
|
||||
heavy Heavy
|
||||
hmq1725 Espers
|
||||
hodl Hodlcoin
|
||||
jha Jackpotcoin
|
||||
keccak Maxcoin
|
||||
keccakc Creative coin
|
||||
lbry LBC, LBRY Credits
|
||||
luffa Luffa
|
||||
lyra2h Hppcoin
|
||||
lyra2re lyra2
|
||||
lyra2rev2 lyra2v2, Vertcoin
|
||||
lyra2z Zcoin (XZC)
|
||||
lyra2z330 Lyra2 330 rows, Zoin (ZOI)
|
||||
m7m Magi (XMG)
|
||||
myr-gr Myriad-Groestl
|
||||
neoscrypt NeoScrypt(128, 2, 1)
|
||||
nist5 Nist5
|
||||
pentablake Pentablake
|
||||
phi1612 phi, LUX coin
|
||||
pluck Pluck:128 (Supcoin)
|
||||
polytimos Ninja
|
||||
quark Quark
|
||||
qubit Qubit
|
||||
scrypt scrypt(1024, 1, 1) (default)
|
||||
scrypt:N scrypt(N, 1, 1)
|
||||
deep Deepcoin (DCN)
|
||||
dmd-gr Diamond-Groestl
|
||||
drop Dropcoin
|
||||
fresh Fresh
|
||||
groestl Groestl coin
|
||||
heavy Heavy
|
||||
hmq1725 Espers
|
||||
hodl Hodlcoin
|
||||
jha Jackpotcoin
|
||||
keccak Maxcoin
|
||||
keccakc Creative coin
|
||||
lbry LBC, LBRY Credits
|
||||
luffa Luffa
|
||||
lyra2h Hppcoin
|
||||
lyra2re lyra2
|
||||
lyra2rev2 lyra2v2, Vertcoin
|
||||
lyra2rev3 lyrav2v3, Vertcoin
|
||||
lyra2z Zcoin (XZC)
|
||||
lyra2z330 Lyra2 330 rows, Zoin (ZOI)
|
||||
m7m Magi (XMG)
|
||||
myr-gr Myriad-Groestl
|
||||
neoscrypt NeoScrypt(128, 2, 1)
|
||||
nist5 Nist5
|
||||
pentablake Pentablake
|
||||
phi1612 phi, LUX coin (original algo)
|
||||
phi2 LUX coin (new algo)
|
||||
pluck Pluck:128 (Supcoin)
|
||||
polytimos Ninja
|
||||
quark Quark
|
||||
qubit Qubit
|
||||
scrypt scrypt(1024, 1, 1) (default)
|
||||
scrypt:N scrypt(N, 1, 1)
|
||||
scryptjane:nf
|
||||
sha256d Double SHA-256
|
||||
sha256t Triple SHA-256, Onecoin (OC)
|
||||
shavite3 Shavite3
|
||||
skein Skein+Sha (Skeincoin)
|
||||
skein2 Double Skein (Woodcoin)
|
||||
skunk Signatum (SIGT)
|
||||
timetravel Machinecoin (MAC)
|
||||
timetravel10 Bitcore
|
||||
tribus Denarius (DNR)
|
||||
vanilla blake256r8vnl (VCash)
|
||||
veltor (VLT)
|
||||
sha256d Double SHA-256
|
||||
sha256t Triple SHA-256, Onecoin (OC)
|
||||
shavite3 Shavite3
|
||||
skein Skein+Sha (Skeincoin)
|
||||
skein2 Double Skein (Woodcoin)
|
||||
skunk Signatum (SIGT)
|
||||
sonoa Sono
|
||||
timetravel Machinecoin (MAC)
|
||||
timetravel10 Bitcore
|
||||
tribus Denarius (DNR)
|
||||
vanilla blake256r8vnl (VCash)
|
||||
veltor (VLT)
|
||||
whirlpool
|
||||
whirlpoolx
|
||||
x11 Dash
|
||||
x11evo Revolvercoin
|
||||
x11gost sib (SibCoin)
|
||||
x12 Galaxie Cash (GCH)
|
||||
x13 X13
|
||||
x13sm3 hsr (Hshare)
|
||||
x14 X14
|
||||
x15 X15
|
||||
x16r Ravencoin
|
||||
x11 Dash
|
||||
x11evo Revolvercoin
|
||||
x11gost sib (SibCoin)
|
||||
x12 Galaxie Cash (GCH)
|
||||
x13 X13
|
||||
x13sm3 hsr (Hshare)
|
||||
x14 X14
|
||||
x15 X15
|
||||
x16r Ravencoin (RVN)
|
||||
x16s pigeoncoin (PGN)
|
||||
x17
|
||||
xevan Bitsend
|
||||
yescrypt Globalboost-Y (BSTY)
|
||||
yescryptr8 BitZeny (ZNY)
|
||||
yescryptr16 Yenten (YTN)
|
||||
yescryptr32 WAVI
|
||||
zr5 Ziftr
|
||||
xevan Bitsend (BSD)
|
||||
yescrypt Globalboost-Y (BSTY)
|
||||
yescryptr8 BitZeny (ZNY)
|
||||
yescryptr16 Eli
|
||||
yescryptr32 WAVI
|
||||
yespower Cryply
|
||||
yespowerr16 Yenten (YTN)
|
||||
zr5 Ziftr
|
||||
|
||||
Errata
|
||||
------
|
||||
|
||||
Cryptonight and variants are no longer supported, use another miner.
|
||||
|
||||
Neoscrypt crashes on Windows, use legacy version.
|
||||
|
||||
AMD CPUs older than Piledriver, including Athlon x2 and Phenom II x4, are not
|
||||
supported by cpuminer-opt due to an incompatible implementation of SSE2 on
|
||||
these CPUs. Some algos may crash the miner with an invalid instruction.
|
||||
|
||||
26
README.txt
26
README.txt
@@ -4,33 +4,35 @@ for Linux and Windows can be found in RELEASE_NOTES.
|
||||
cpuminer is a console program that is executed from a DOS command prompt.
|
||||
There is no GUI and no mouse support.
|
||||
|
||||
Miner programs are often flagged as malware by antivirus programs. This is
|
||||
a false positive, they are flagged simply because they are cryptocurrency
|
||||
miners. The source code is open for anyone to inspect. If you don't trust
|
||||
the software, don't use it.
|
||||
|
||||
Choose the exe that best matches you CPU's features or use trial and
|
||||
error to find the fastest one that doesn't crash. Pay attention to
|
||||
the features listed at cpuminer startup to ensure you are mining at
|
||||
optimum speed using all the available features.
|
||||
optimum speed using the best available features.
|
||||
|
||||
Architecture names and compile options used are only provided for Intel
|
||||
Core series. Pentium and Celeron often have fewer features.
|
||||
Core series. Even the newest Pentium and Celeron CPUs are often missing
|
||||
features.
|
||||
|
||||
AMD CPUs older than Piledriver, including Athlon x2 and Phenom II x4, are not
|
||||
supported by cpuminer-opt due to an incompatible implementation of SSE2 on
|
||||
these CPUs. Some algos may crash the miner with an invalid instruction.
|
||||
Users are recommended to use an unoptimized miner such as cpuminer-multi.
|
||||
|
||||
Exe name Compile flags Arch name
|
||||
Exe name Compile flags Arch name
|
||||
|
||||
cpuminer-sse2.exe "-march=core2" Core2, Nehalem
|
||||
cpuminer-aes-sse42.exe "-maes -msse4.2" Westmere
|
||||
cpuminer-aes-avx.exe "-march=corei7-avx" Sandybridge, Ivybridge
|
||||
cpuminer-avx2.exe "-march=core-avx2" Haswell...
|
||||
cpuminer-avx2-sha.exe "-march=core-avx2 -msha" Ryzen
|
||||
cpuminer-sse2.exe "-msse2" Core2, Nehalem
|
||||
cpuminer-aes-sse42.exe "-march=westmere" Westmere
|
||||
cpuminer-avx.exe "-march=corei7-avx" Sandy-Ivybridge
|
||||
cpuminer-avx2.exe "-march=core-avx2" Haswell, Sky-Kaby-Coffeelake
|
||||
cpuminer-zen "-march=znver1" AMD Ryzen, Threadripper
|
||||
|
||||
If you like this software feel free to donate:
|
||||
|
||||
BTC: 12tdvfF7KmAsihBXQXynT6E6th2c2pByTT
|
||||
ETH: 0x72122edabcae9d3f57eab0729305a425f6fef6d0
|
||||
LTC: LdUwoHJnux9r9EKqFWNvAi45kQompHk6e8
|
||||
BCH: 1QKYkB6atn4P7RFozyziAXLEnurwnUM1cQ
|
||||
BTG: GVUyECtRHeC5D58z9F3nGGfVQndwnsPnHQ
|
||||
|
||||
|
||||
|
||||
304
RELEASE_NOTES
304
RELEASE_NOTES
@@ -1,11 +1,11 @@
|
||||
puminer-opt now supports HW SHA acceleration available on AMD Ryzen CPUs.
|
||||
cpuminer-opt is a console program run from the command line using the
|
||||
keyboard, not the mouse.
|
||||
|
||||
cpuminer-opt now supports HW SHA acceleration available on AMD Ryzen CPUs.
|
||||
This feature requires recent SW including GCC version 5 or higher and
|
||||
openssl version 1.1 or higher. It may also require using "-march=znver1"
|
||||
compile flag.
|
||||
|
||||
HW SHA support is only available when compiled from source, Windows binaries
|
||||
are not yet available.
|
||||
|
||||
cpuminer-opt is a console program, if you're using a mouse you're doing it
|
||||
wrong.
|
||||
|
||||
@@ -13,11 +13,11 @@ Security warning
|
||||
----------------
|
||||
|
||||
Miner programs are often flagged as malware by antivirus programs. This is
|
||||
a false positive, they are flagged simply because they are miners. The source
|
||||
code is open for anyone to inspect. If you don't trust the software, don't use
|
||||
it.
|
||||
a false positive, they are flagged simply because they are cryptocurrency
|
||||
miners. The source code is open for anyone to inspect. If you don't trust
|
||||
the software, don't use it.
|
||||
|
||||
The cryptographic code has been taken from trusted sources but has been
|
||||
The cryptographic hashing code has been taken from trusted sources but has been
|
||||
modified for speed at the expense of accepted security practices. This
|
||||
code should not be imported into applications where secure cryptography is
|
||||
required.
|
||||
@@ -25,141 +25,174 @@ required.
|
||||
Compile Instructions
|
||||
--------------------
|
||||
|
||||
Requirements:
|
||||
See INSTALL_LINUX or INSTALL_WINDOWS fror compile instruuctions
|
||||
|
||||
Requirements
|
||||
------------
|
||||
|
||||
Intel Core2 or newer, or AMD Steamroller or newer CPU. ARM CPUs are not
|
||||
supported.
|
||||
64 bit Linux or Windows operating system. Apple is not supported.
|
||||
|
||||
Building on linux prerequisites:
|
||||
|
||||
It is assumed users know how to install packages on their system and
|
||||
be able to compile standard source packages. This is basic Linux and
|
||||
beyond the scope of cpuminer-opt.
|
||||
|
||||
Make sure you have the basic development packages installed.
|
||||
Here is a good start:
|
||||
|
||||
http://askubuntu.com/questions/457526/how-to-install-cpuminer-in-ubuntu
|
||||
|
||||
Install any additional dependencies needed by cpuminer-opt. The list below
|
||||
are some of the ones that may not be in the default install and need to
|
||||
be installed manually. There may be others, read the error messages they
|
||||
will give a clue as to the missing package.
|
||||
|
||||
The following command should install everything you need on Debian based
|
||||
distributions such as Ubuntu:
|
||||
|
||||
sudo apt-get install build-essential libssl-dev libcurl4-openssl-dev libjansson-dev libgmp-dev automake
|
||||
|
||||
|
||||
build-essential (for Ubuntu, Development Tools package group on Fedora)
|
||||
automake
|
||||
libjansson-dev
|
||||
libgmp-dev
|
||||
libcurl4-openssl-dev
|
||||
libssl-dev
|
||||
pthreads
|
||||
zlib
|
||||
|
||||
SHA support on AMD Ryzen CPUs requires gcc version 5 or higher and openssl 1.1
|
||||
or higher. Reports of improved performiance on Ryzen when using openssl 1.0.2
|
||||
have been due to AVX and AVX2 optimizations added to that version.
|
||||
Additional improvements are expected on Ryzen with openssl 1.1.
|
||||
"-march-znver1" or "-msha".
|
||||
|
||||
Additional instructions for static compilalation can be found here:
|
||||
https://lxadm.com/Static_compilation_of_cpuminer
|
||||
Static builds should only considered in a homogeneous HW and SW environment.
|
||||
Local builds will always have the best performance and compatibility.
|
||||
|
||||
Extract cpuminer source.
|
||||
|
||||
tar xvzf cpuminer-opt-x.y.z.tar.gz
|
||||
cd cpuminer-opt-x.y.z
|
||||
|
||||
Run ./build.sh to build on Linux or execute the following commands.
|
||||
|
||||
./autogen.sh
|
||||
CFLAGS="-O3 -march=native -Wall" CXXFLAGS="$CFLAGS -std=gnu++11" ./configure --with-curl
|
||||
make
|
||||
|
||||
Additional optional compile flags, add the following to CFLAGS to activate:
|
||||
|
||||
-DUSE_SPH_SHA
|
||||
|
||||
SPH may give slightly better performance on algos that use sha256 when using
|
||||
openssl 1.0.1 or older. Openssl 1.0.2 adds AVX2 and 1.1 adds SHA and perform
|
||||
better than SPH. This option is ignored when 4-way is used, even for CPUs
|
||||
with SHA.
|
||||
|
||||
Start mining.
|
||||
|
||||
./cpuminer -a algo -o url -u username -p password
|
||||
|
||||
Windows
|
||||
|
||||
Precompiled Windows binaries are built on a Linux host using Mingw
|
||||
with a more recent compiler than the following Windows hosted procedure.
|
||||
|
||||
Building on Windows prerequisites:
|
||||
|
||||
msys
|
||||
mingw_w64
|
||||
Visual C++ redistributable 2008 X64
|
||||
openssl
|
||||
|
||||
Install msys and mingw_w64, only needed once.
|
||||
|
||||
Unpack msys into C:\msys or your preferred directory.
|
||||
|
||||
Install mingw_w64 from win-builds.
|
||||
Follow instructions, check "msys or cygwin" and "x86_64" and accept default
|
||||
existing msys instalation.
|
||||
|
||||
Open a msys shell by double clicking on msys.bat.
|
||||
Note that msys shell uses linux syntax for file specifications, "C:\" is
|
||||
mounted at "/c/".
|
||||
|
||||
Add mingw bin directory to PATH variable
|
||||
PATH="/c/msys/opt/windows_64/bin/:$PATH"
|
||||
|
||||
Instalation complete, compile cpuminer-opt.
|
||||
|
||||
Unpack cpuminer-opt source files using tar from msys shell, or using 7zip
|
||||
or similar Windows program.
|
||||
|
||||
In msys shell cd to miner directory.
|
||||
cd /c/path/to/cpuminer-opt
|
||||
|
||||
Run build.sh to build on Windows or execute the following commands.
|
||||
|
||||
./autogen.sh
|
||||
CFLAGS="-O3 -march=native -Wall" ./configure --with-curl
|
||||
make
|
||||
|
||||
Start mining
|
||||
|
||||
cpuminer.exe -a algo -o url -u user -p password
|
||||
|
||||
The following tips may be useful for older AMD CPUs.
|
||||
|
||||
AMD CPUs older than Steamroller, including Athlon x2 and Phenom II x4, are
|
||||
not supported by cpuminer-opt due to an incompatible implementation of SSE2
|
||||
on these CPUs. Some algos may crash the miner with an invalid instruction.
|
||||
Users are recommended to use an unoptimized miner such as cpuminer-multi.
|
||||
|
||||
Some users with AMD CPUs without AES_NI have reported problems compiling
|
||||
with build.sh or "-march=native". Problems have included compile errors
|
||||
and poor performance. These users are recommended to compile manually
|
||||
specifying "-march=btver1" on the configure command line.
|
||||
|
||||
Support for even older x86_64 without AES_NI or SSE2 is not availble.
|
||||
|
||||
64 bit Linux or Windows operating system. Apple and Android are not supported.
|
||||
|
||||
Change Log
|
||||
----------
|
||||
|
||||
v3.9.4
|
||||
|
||||
Faster AVX2 for lyra2v3, quark, anime.
|
||||
Fixed skein AVX2 regression (invalid shares since v3.9.0) and faster.
|
||||
Faster skein2 with 4way AVX2 enabled.
|
||||
Automatic SHA override on Ryzen CPUs, no need for -DRYZEN compile flag.
|
||||
Ongoing restructuring.
|
||||
|
||||
v3.9.3.1
|
||||
|
||||
Skipped v3.9.3 due to misidentification of v3.9.2.5 as v3.9.3.
|
||||
Fixed x16r algo 25% invalid share reject rate. The bug may have also
|
||||
affected other algos.
|
||||
|
||||
v3.9.2.5
|
||||
|
||||
Fixed 2 regressions: hodl AES detection, x16r invalid shares with AVX2.
|
||||
More restructuring.
|
||||
|
||||
v3.9.2.4
|
||||
|
||||
Yet another affinity fix. Hopefully the last one.
|
||||
|
||||
v3.9.2.3
|
||||
|
||||
Another cpu-affinity fix.
|
||||
Disabled test code that fails to compile on some CPUs with limited
|
||||
AVX512 capabilities.
|
||||
|
||||
v3.9.2.2
|
||||
|
||||
Fixed some day one cpu-affinity issues.
|
||||
|
||||
v3.9.2
|
||||
|
||||
Added sha256q algo.
|
||||
Yespower now uses openssl SHA256, but no observable hash rate increase
|
||||
on Ryzen.
|
||||
Ongoing rearchitecting.
|
||||
Lyra2z now hashes 8-way on CPUs with AVX2.
|
||||
Lyra2 (all including phi2) now runs optimized code with SSE2.
|
||||
|
||||
v3.9.1.1
|
||||
|
||||
Fixed lyra2v3 AVX and below.
|
||||
|
||||
Compiling on Windows using Cygwin now works. Simply use "./build.sh"
|
||||
just like on Linux. It isn't portable therefore the binaries package will
|
||||
continue to use the existing procedure.
|
||||
The Cygwin procedure will be documented in more detail later and will
|
||||
include a list of packages that need to be installed.
|
||||
|
||||
v3.9.1
|
||||
|
||||
Fixed AVX2 version of anime algo.
|
||||
|
||||
Added sonoa algo.
|
||||
|
||||
Added "-DRYZEN_" compile option for Ryzen to override 4-way hashing when algo
|
||||
contains sha256 and use SHA instead. This is due to a combination of
|
||||
the introduction of HW SHA support combined with the poor performance
|
||||
of AVX2 on Ryzen. The Windows binaries package replaces cpuminer-avx2-sha
|
||||
with cpuminer-zen compiled with the override. Refer to the build instructions
|
||||
for more information.
|
||||
|
||||
Ongoing restructuring to streamline the process, reduce latency,
|
||||
reduce memory usage and unnecessary copying of data. Most of these
|
||||
will not result in a notoceably higher reported hashrate as the
|
||||
change simply reduces the time wasted that wasn't factored into the
|
||||
hash rate reported by the miner. In short, less dead time resulting in
|
||||
a higher net hashrate.
|
||||
|
||||
One of these measures to reduce latency also results in an enhanced
|
||||
share submission message including the share number*, the CPU thread,
|
||||
and the vector lane that found the solution. The time difference between
|
||||
the share submission and acceptance (or rejection) response indicates
|
||||
network ltatency. One other effect of this change is a reduction in hash
|
||||
meter messages because the scan function no longer exits when a share is
|
||||
found. Scan cycles will go longer and submit multiple shares per cycle.
|
||||
*the share number is antcipated and includes both accepted and rejected
|
||||
shares. Because the share is antipated and not synchronized it may be
|
||||
incorrect in time of very rapid share submission. Under most conditions
|
||||
it should be easy to match the submission with the corresponding response.
|
||||
|
||||
Removed "-DUSE_SPH_SHA" option, all users should have a recent version of
|
||||
openssl installed: v1.0.2 (Ubuntu 16.04) or better. Ryzen SHA requires
|
||||
v1.1.0 or better. Ryzen SHA is not used when hashing multi-way parallel.
|
||||
Ryzen SHA is available in the Windows binaries release package.
|
||||
|
||||
Improved compile instructions, now in seperate files: INSTALL_LINUX and
|
||||
INSTALL_WINDOWS. The Windows instructions are used to build the binaries
|
||||
release package. It's built on a Linux system either running as a virtual
|
||||
machine or a seperate computer. At this time there is no known way to
|
||||
build natively on a Windows system.
|
||||
|
||||
v3.9.0.1
|
||||
|
||||
Isolate Windows CPU groups code when CPU groups support not explicitly defined.
|
||||
|
||||
v3.9.0
|
||||
|
||||
Added support for Windows CPU groups.
|
||||
Fixed BIP34 coinbase height.
|
||||
Prep work for AVX512.
|
||||
Added lyra2rev3 for the vertcoin algo change.
|
||||
Added yespower, yespowerr16 (Yenten)
|
||||
Added phi2 algo for LUX
|
||||
Discontinued support for cryptonight and variants.
|
||||
|
||||
v3.8.8.1
|
||||
|
||||
Fixed x16r.
|
||||
Removed cryptonight variant check due to false positives.
|
||||
API displays hashrate before shares are submitted.
|
||||
|
||||
v3.8.8
|
||||
|
||||
Added cryptonightv7 for Monero.
|
||||
|
||||
v3.8.7.2
|
||||
|
||||
Fixed argon2d-dyn regression in v3.8.7.1.
|
||||
Changed compile options for aes-sse42 Windows build to -march=westmere
|
||||
|
||||
v3.8.7.1
|
||||
|
||||
Fixed argon2d-uis low difficulty rejects.
|
||||
Fixed argon2d aliases.
|
||||
|
||||
v3.8.7
|
||||
|
||||
Added argon2d4096 (alias argon2d-uis) for Unitus (UIS).
|
||||
argon2d-crds and argon2d-dyn renamed to argon2d250 and argon2d500 respectively.
|
||||
The old names are recognized as aliases.
|
||||
AVX512 is now supported for argon2d algos, Linux only.
|
||||
AVX is no longer a reported feature and an AVX Windows binary is no longer
|
||||
provided. Use AES-SSE42 build instead.
|
||||
|
||||
v3.8.6.1
|
||||
|
||||
Faster argon2d* AVX2.
|
||||
Untested AVX-512 for argon2d*, YMMV.
|
||||
|
||||
v3.8.6
|
||||
|
||||
Fixed argon2 regression in v3.8.5.
|
||||
Added x16s algo for Pigeoncoin.
|
||||
Some code cleanup.
|
||||
|
||||
v3.8.5
|
||||
|
||||
Added argon2d-crds and argon2d-dyn algos.
|
||||
sha256t 8 way AVX2 & 4 way SSE4.2 optimized.
|
||||
CPUs with SSE4.2 get optimizations previously reserved for AVX.
|
||||
|
||||
v3.8.4.1
|
||||
|
||||
Fixed sha256t low difficulty rejects.
|
||||
@@ -296,6 +329,7 @@ Changed default sha256 and sha512 to openssl. This should be used when
|
||||
compiling with openssl 1.0.2 or higher (Ubuntu 16.04).
|
||||
This should increase the hashrate for yescrypt, yescryptr16, m7m, xevan, skein,
|
||||
myr-gr & others when openssl 1.0.2 is installed.
|
||||
Note: -DUSE_SPH_SHA has been removed in v3.9.1.
|
||||
Users with openssl 1.0.1 (Ubuntu 14.04) may get better perforance by adding
|
||||
"-DUSE_SPH_SHA" to CLAGS.
|
||||
Windows binaries are compiled with -DUSE_SPH_SHA and won't get the speedup.
|
||||
|
||||
69
aclocal.m4
vendored
69
aclocal.m4
vendored
@@ -1,6 +1,6 @@
|
||||
# generated automatically by aclocal 1.14.1 -*- Autoconf -*-
|
||||
# generated automatically by aclocal 1.15.1 -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 1996-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 1996-2017 Free Software Foundation, Inc.
|
||||
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -20,7 +20,7 @@ You have another version of autoconf. It may work, but is not guaranteed to.
|
||||
If you have problems, you may need to regenerate the build system entirely.
|
||||
To do so, use the procedure documented by the package, typically 'autoreconf'.])])
|
||||
|
||||
# Copyright (C) 2002-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2002-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -32,10 +32,10 @@ To do so, use the procedure documented by the package, typically 'autoreconf'.])
|
||||
# generated from the m4 files accompanying Automake X.Y.
|
||||
# (This private macro should not be called outside this file.)
|
||||
AC_DEFUN([AM_AUTOMAKE_VERSION],
|
||||
[am__api_version='1.14'
|
||||
[am__api_version='1.15'
|
||||
dnl Some users find AM_AUTOMAKE_VERSION and mistake it for a way to
|
||||
dnl require some minimum version. Point them to the right macro.
|
||||
m4_if([$1], [1.14.1], [],
|
||||
m4_if([$1], [1.15.1], [],
|
||||
[AC_FATAL([Do not call $0, use AM_INIT_AUTOMAKE([$1]).])])dnl
|
||||
])
|
||||
|
||||
@@ -51,14 +51,14 @@ m4_define([_AM_AUTOCONF_VERSION], [])
|
||||
# Call AM_AUTOMAKE_VERSION and AM_AUTOMAKE_VERSION so they can be traced.
|
||||
# This function is AC_REQUIREd by AM_INIT_AUTOMAKE.
|
||||
AC_DEFUN([AM_SET_CURRENT_AUTOMAKE_VERSION],
|
||||
[AM_AUTOMAKE_VERSION([1.14.1])dnl
|
||||
[AM_AUTOMAKE_VERSION([1.15.1])dnl
|
||||
m4_ifndef([AC_AUTOCONF_VERSION],
|
||||
[m4_copy([m4_PACKAGE_VERSION], [AC_AUTOCONF_VERSION])])dnl
|
||||
_AM_AUTOCONF_VERSION(m4_defn([AC_AUTOCONF_VERSION]))])
|
||||
|
||||
# Figure out how to run the assembler. -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 2001-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2001-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -78,7 +78,7 @@ _AM_IF_OPTION([no-dependencies],, [_AM_DEPENDENCIES([CCAS])])dnl
|
||||
|
||||
# AM_AUX_DIR_EXPAND -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 2001-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2001-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -123,15 +123,14 @@ _AM_IF_OPTION([no-dependencies],, [_AM_DEPENDENCIES([CCAS])])dnl
|
||||
# configured tree to be moved without reconfiguration.
|
||||
|
||||
AC_DEFUN([AM_AUX_DIR_EXPAND],
|
||||
[dnl Rely on autoconf to set up CDPATH properly.
|
||||
AC_PREREQ([2.50])dnl
|
||||
# expand $ac_aux_dir to an absolute path
|
||||
am_aux_dir=`cd $ac_aux_dir && pwd`
|
||||
[AC_REQUIRE([AC_CONFIG_AUX_DIR_DEFAULT])dnl
|
||||
# Expand $ac_aux_dir to an absolute path.
|
||||
am_aux_dir=`cd "$ac_aux_dir" && pwd`
|
||||
])
|
||||
|
||||
# AM_CONDITIONAL -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 1997-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 1997-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -162,7 +161,7 @@ AC_CONFIG_COMMANDS_PRE(
|
||||
Usually this means the macro was only invoked conditionally.]])
|
||||
fi])])
|
||||
|
||||
# Copyright (C) 1999-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 1999-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -353,7 +352,7 @@ _AM_SUBST_NOTMAKE([am__nodep])dnl
|
||||
|
||||
# Generate code to set up dependency tracking. -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 1999-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 1999-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -429,7 +428,7 @@ AC_DEFUN([AM_OUTPUT_DEPENDENCY_COMMANDS],
|
||||
|
||||
# Do all the work for Automake. -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 1996-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 1996-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -519,8 +518,8 @@ AC_REQUIRE([AC_PROG_MKDIR_P])dnl
|
||||
# <http://lists.gnu.org/archive/html/automake/2012-07/msg00001.html>
|
||||
# <http://lists.gnu.org/archive/html/automake/2012-07/msg00014.html>
|
||||
AC_SUBST([mkdir_p], ['$(MKDIR_P)'])
|
||||
# We need awk for the "check" target. The system "awk" is bad on
|
||||
# some platforms.
|
||||
# We need awk for the "check" target (and possibly the TAP driver). The
|
||||
# system "awk" is bad on some platforms.
|
||||
AC_REQUIRE([AC_PROG_AWK])dnl
|
||||
AC_REQUIRE([AC_PROG_MAKE_SET])dnl
|
||||
AC_REQUIRE([AM_SET_LEADING_DOT])dnl
|
||||
@@ -593,7 +592,11 @@ to "yes", and re-run configure.
|
||||
END
|
||||
AC_MSG_ERROR([Your 'rm' program is bad, sorry.])
|
||||
fi
|
||||
fi])
|
||||
fi
|
||||
dnl The trailing newline in this macro's definition is deliberate, for
|
||||
dnl backward compatibility and to allow trailing 'dnl'-style comments
|
||||
dnl after the AM_INIT_AUTOMAKE invocation. See automake bug#16841.
|
||||
])
|
||||
|
||||
dnl Hook into '_AC_COMPILER_EXEEXT' early to learn its expansion. Do not
|
||||
dnl add the conditional right here, as _AC_COMPILER_EXEEXT may be further
|
||||
@@ -622,7 +625,7 @@ for _am_header in $config_headers :; do
|
||||
done
|
||||
echo "timestamp for $_am_arg" >`AS_DIRNAME(["$_am_arg"])`/stamp-h[]$_am_stamp_count])
|
||||
|
||||
# Copyright (C) 2001-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2001-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -633,7 +636,7 @@ echo "timestamp for $_am_arg" >`AS_DIRNAME(["$_am_arg"])`/stamp-h[]$_am_stamp_co
|
||||
# Define $install_sh.
|
||||
AC_DEFUN([AM_PROG_INSTALL_SH],
|
||||
[AC_REQUIRE([AM_AUX_DIR_EXPAND])dnl
|
||||
if test x"${install_sh}" != xset; then
|
||||
if test x"${install_sh+set}" != xset; then
|
||||
case $am_aux_dir in
|
||||
*\ * | *\ *)
|
||||
install_sh="\${SHELL} '$am_aux_dir/install-sh'" ;;
|
||||
@@ -643,7 +646,7 @@ if test x"${install_sh}" != xset; then
|
||||
fi
|
||||
AC_SUBST([install_sh])])
|
||||
|
||||
# Copyright (C) 2003-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2003-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -665,7 +668,7 @@ AC_SUBST([am__leading_dot])])
|
||||
# Add --enable-maintainer-mode option to configure. -*- Autoconf -*-
|
||||
# From Jim Meyering
|
||||
|
||||
# Copyright (C) 1996-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 1996-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -700,7 +703,7 @@ AC_MSG_CHECKING([whether to enable maintainer-specific portions of Makefiles])
|
||||
|
||||
# Check to see how 'make' treats includes. -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 2001-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2001-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -750,7 +753,7 @@ rm -f confinc confmf
|
||||
|
||||
# Fake the existence of programs that GNU maintainers use. -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 1997-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 1997-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -789,7 +792,7 @@ fi
|
||||
|
||||
# Helper functions for option handling. -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 2001-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2001-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -818,7 +821,7 @@ AC_DEFUN([_AM_SET_OPTIONS],
|
||||
AC_DEFUN([_AM_IF_OPTION],
|
||||
[m4_ifset(_AM_MANGLE_OPTION([$1]), [$2], [$3])])
|
||||
|
||||
# Copyright (C) 1999-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 1999-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -865,7 +868,7 @@ AC_LANG_POP([C])])
|
||||
# For backward compatibility.
|
||||
AC_DEFUN_ONCE([AM_PROG_CC_C_O], [AC_REQUIRE([AC_PROG_CC])])
|
||||
|
||||
# Copyright (C) 2001-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2001-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -884,7 +887,7 @@ AC_DEFUN([AM_RUN_LOG],
|
||||
|
||||
# Check to make sure that the build environment is sane. -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 1996-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 1996-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -965,7 +968,7 @@ AC_CONFIG_COMMANDS_PRE(
|
||||
rm -f conftest.file
|
||||
])
|
||||
|
||||
# Copyright (C) 2009-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2009-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -1025,7 +1028,7 @@ AC_SUBST([AM_BACKSLASH])dnl
|
||||
_AM_SUBST_NOTMAKE([AM_BACKSLASH])dnl
|
||||
])
|
||||
|
||||
# Copyright (C) 2001-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2001-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -1053,7 +1056,7 @@ fi
|
||||
INSTALL_STRIP_PROGRAM="\$(install_sh) -c -s"
|
||||
AC_SUBST([INSTALL_STRIP_PROGRAM])])
|
||||
|
||||
# Copyright (C) 2006-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2006-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
@@ -1072,7 +1075,7 @@ AC_DEFUN([AM_SUBST_NOTMAKE], [_AM_SUBST_NOTMAKE($@)])
|
||||
|
||||
# Check how to create a tarball. -*- Autoconf -*-
|
||||
|
||||
# Copyright (C) 2004-2013 Free Software Foundation, Inc.
|
||||
# Copyright (C) 2004-2017 Free Software Foundation, Inc.
|
||||
#
|
||||
# This file is free software; the Free Software Foundation
|
||||
# gives unlimited permission to copy and/or distribute it,
|
||||
|
||||
202
algo-gate-api.c
202
algo-gate-api.c
@@ -69,6 +69,8 @@ void do_nothing () {}
|
||||
bool return_true () { return true; }
|
||||
bool return_false () { return false; }
|
||||
void *return_null () { return NULL; }
|
||||
void call_error () { printf("ERR: Uninitialized function pointer\n"); }
|
||||
|
||||
|
||||
void algo_not_tested()
|
||||
{
|
||||
@@ -113,7 +115,8 @@ void init_algo_gate( algo_gate_t* gate )
|
||||
gate->hash_suw = (void*)&null_hash_suw;
|
||||
gate->get_new_work = (void*)&std_get_new_work;
|
||||
gate->get_nonceptr = (void*)&std_get_nonceptr;
|
||||
gate->display_extra_data = (void*)&do_nothing;
|
||||
gate->work_decode = (void*)&std_le_work_decode;
|
||||
gate->decode_extra_data = (void*)&do_nothing;
|
||||
gate->wait_for_diff = (void*)&std_wait_for_diff;
|
||||
gate->get_max64 = (void*)&get_max64_0x1fffffLL;
|
||||
gate->gen_merkle_root = (void*)&sha256d_gen_merkle_root;
|
||||
@@ -121,7 +124,6 @@ void init_algo_gate( algo_gate_t* gate )
|
||||
gate->build_stratum_request = (void*)&std_le_build_stratum_request;
|
||||
gate->malloc_txs_request = (void*)&std_malloc_txs_request;
|
||||
gate->set_target = (void*)&std_set_target;
|
||||
gate->work_decode = (void*)&std_le_work_decode;
|
||||
gate->submit_getwork_result = (void*)&std_le_submit_getwork_result;
|
||||
gate->build_block_header = (void*)&std_build_block_header;
|
||||
gate->build_extraheader = (void*)&std_build_extraheader;
|
||||
@@ -132,11 +134,11 @@ void init_algo_gate( algo_gate_t* gate )
|
||||
gate->do_this_thread = (void*)&return_true;
|
||||
gate->longpoll_rpc_call = (void*)&std_longpoll_rpc_call;
|
||||
gate->stratum_handle_response = (void*)&std_stratum_handle_response;
|
||||
gate->get_work_data_size = (void*)&std_get_work_data_size;
|
||||
gate->optimizations = EMPTY_SET;
|
||||
gate->ntime_index = STD_NTIME_INDEX;
|
||||
gate->nbits_index = STD_NBITS_INDEX;
|
||||
gate->nonce_index = STD_NONCE_INDEX;
|
||||
gate->work_data_size = STD_WORK_DATA_SIZE;
|
||||
gate->work_cmp_size = STD_WORK_CMP_SIZE;
|
||||
}
|
||||
|
||||
@@ -157,78 +159,95 @@ bool register_algo_gate( int algo, algo_gate_t *gate )
|
||||
|
||||
switch (algo)
|
||||
{
|
||||
case ALGO_ALLIUM: register_allium_algo ( gate ); break;
|
||||
case ALGO_ANIME: register_anime_algo ( gate ); break;
|
||||
case ALGO_ARGON2: register_argon2_algo ( gate ); break;
|
||||
case ALGO_AXIOM: register_axiom_algo ( gate ); break;
|
||||
case ALGO_BASTION: register_bastion_algo ( gate ); break;
|
||||
case ALGO_BLAKE: register_blake_algo ( gate ); break;
|
||||
case ALGO_BLAKECOIN: register_blakecoin_algo ( gate ); break;
|
||||
case ALGO_ALLIUM: register_allium_algo ( gate ); break;
|
||||
case ALGO_ANIME: register_anime_algo ( gate ); break;
|
||||
case ALGO_ARGON2: register_argon2_algo ( gate ); break;
|
||||
case ALGO_ARGON2D250: register_argon2d_crds_algo ( gate ); break;
|
||||
case ALGO_ARGON2D500: register_argon2d_dyn_algo ( gate ); break;
|
||||
case ALGO_ARGON2D4096: register_argon2d4096_algo ( gate ); break;
|
||||
case ALGO_AXIOM: register_axiom_algo ( gate ); break;
|
||||
case ALGO_BASTION: register_bastion_algo ( gate ); break;
|
||||
case ALGO_BLAKE: register_blake_algo ( gate ); break;
|
||||
case ALGO_BLAKECOIN: register_blakecoin_algo ( gate ); break;
|
||||
// case ALGO_BLAKE2B: register_blake2b_algo ( gate ); break;
|
||||
case ALGO_BLAKE2S: register_blake2s_algo ( gate ); break;
|
||||
case ALGO_C11: register_c11_algo ( gate ); break;
|
||||
case ALGO_CRYPTOLIGHT: register_cryptolight_algo ( gate ); break;
|
||||
case ALGO_CRYPTONIGHT: register_cryptonight_algo ( gate ); break;
|
||||
case ALGO_DECRED: register_decred_algo ( gate ); break;
|
||||
case ALGO_DEEP: register_deep_algo ( gate ); break;
|
||||
case ALGO_DMD_GR: register_dmd_gr_algo ( gate ); break;
|
||||
case ALGO_DROP: register_drop_algo ( gate ); break;
|
||||
case ALGO_FRESH: register_fresh_algo ( gate ); break;
|
||||
case ALGO_GROESTL: register_groestl_algo ( gate ); break;
|
||||
case ALGO_HEAVY: register_heavy_algo ( gate ); break;
|
||||
case ALGO_HMQ1725: register_hmq1725_algo ( gate ); break;
|
||||
case ALGO_HODL: register_hodl_algo ( gate ); break;
|
||||
case ALGO_JHA: register_jha_algo ( gate ); break;
|
||||
case ALGO_KECCAK: register_keccak_algo ( gate ); break;
|
||||
case ALGO_KECCAKC: register_keccakc_algo ( gate ); break;
|
||||
case ALGO_LBRY: register_lbry_algo ( gate ); break;
|
||||
case ALGO_LUFFA: register_luffa_algo ( gate ); break;
|
||||
case ALGO_LYRA2H: register_lyra2h_algo ( gate ); break;
|
||||
case ALGO_LYRA2RE: register_lyra2re_algo ( gate ); break;
|
||||
case ALGO_LYRA2REV2: register_lyra2rev2_algo ( gate ); break;
|
||||
case ALGO_LYRA2Z: register_lyra2z_algo ( gate ); break;
|
||||
case ALGO_LYRA2Z330: register_lyra2z330_algo ( gate ); break;
|
||||
case ALGO_M7M: register_m7m_algo ( gate ); break;
|
||||
case ALGO_MYR_GR: register_myriad_algo ( gate ); break;
|
||||
case ALGO_NEOSCRYPT: register_neoscrypt_algo ( gate ); break;
|
||||
case ALGO_NIST5: register_nist5_algo ( gate ); break;
|
||||
case ALGO_PENTABLAKE: register_pentablake_algo ( gate ); break;
|
||||
case ALGO_PHI1612: register_phi1612_algo ( gate ); break;
|
||||
case ALGO_PLUCK: register_pluck_algo ( gate ); break;
|
||||
case ALGO_POLYTIMOS: register_polytimos_algo ( gate ); break;
|
||||
case ALGO_QUARK: register_quark_algo ( gate ); break;
|
||||
case ALGO_QUBIT: register_qubit_algo ( gate ); break;
|
||||
case ALGO_SCRYPT: register_scrypt_algo ( gate ); break;
|
||||
case ALGO_SCRYPTJANE: register_scryptjane_algo ( gate ); break;
|
||||
case ALGO_SHA256D: register_sha256d_algo ( gate ); break;
|
||||
case ALGO_SHA256T: register_sha256t_algo ( gate ); break;
|
||||
case ALGO_SHAVITE3: register_shavite_algo ( gate ); break;
|
||||
case ALGO_SKEIN: register_skein_algo ( gate ); break;
|
||||
case ALGO_SKEIN2: register_skein2_algo ( gate ); break;
|
||||
case ALGO_SKUNK: register_skunk_algo ( gate ); break;
|
||||
case ALGO_TIMETRAVEL: register_timetravel_algo ( gate ); break;
|
||||
case ALGO_TIMETRAVEL10: register_timetravel10_algo( gate ); break;
|
||||
case ALGO_TRIBUS: register_tribus_algo ( gate ); break;
|
||||
case ALGO_VANILLA: register_vanilla_algo ( gate ); break;
|
||||
case ALGO_VELTOR: register_veltor_algo ( gate ); break;
|
||||
case ALGO_WHIRLPOOL: register_whirlpool_algo ( gate ); break;
|
||||
case ALGO_WHIRLPOOLX: register_whirlpoolx_algo ( gate ); break;
|
||||
case ALGO_X11: register_x11_algo ( gate ); break;
|
||||
case ALGO_X11EVO: register_x11evo_algo ( gate ); break;
|
||||
case ALGO_X11GOST: register_x11gost_algo ( gate ); break;
|
||||
case ALGO_X12: register_x12_algo ( gate ); break;
|
||||
case ALGO_X13: register_x13_algo ( gate ); break;
|
||||
case ALGO_X13SM3: register_x13sm3_algo ( gate ); break;
|
||||
case ALGO_X14: register_x14_algo ( gate ); break;
|
||||
case ALGO_X15: register_x15_algo ( gate ); break;
|
||||
case ALGO_X16R: register_x16r_algo ( gate ); break;
|
||||
case ALGO_X17: register_x17_algo ( gate ); break;
|
||||
case ALGO_XEVAN: register_xevan_algo ( gate ); break;
|
||||
case ALGO_YESCRYPT: register_yescrypt_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR8: register_yescryptr8_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR16: register_yescryptr16_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR32: register_yescryptr32_algo ( gate ); break;
|
||||
case ALGO_ZR5: register_zr5_algo ( gate ); break;
|
||||
case ALGO_BLAKE2S: register_blake2s_algo ( gate ); break;
|
||||
case ALGO_C11: register_c11_algo ( gate ); break;
|
||||
case ALGO_CRYPTOLIGHT: register_cryptolight_algo ( gate ); break;
|
||||
case ALGO_CRYPTONIGHT: register_cryptonight_algo ( gate ); break;
|
||||
case ALGO_CRYPTONIGHTV7:register_cryptonightv7_algo( gate ); break;
|
||||
case ALGO_DECRED: register_decred_algo ( gate ); break;
|
||||
case ALGO_DEEP: register_deep_algo ( gate ); break;
|
||||
case ALGO_DMD_GR: register_dmd_gr_algo ( gate ); break;
|
||||
case ALGO_DROP: register_drop_algo ( gate ); break;
|
||||
case ALGO_FRESH: register_fresh_algo ( gate ); break;
|
||||
case ALGO_GROESTL: register_groestl_algo ( gate ); break;
|
||||
case ALGO_HEAVY: register_heavy_algo ( gate ); break;
|
||||
case ALGO_HMQ1725: register_hmq1725_algo ( gate ); break;
|
||||
case ALGO_HODL: register_hodl_algo ( gate ); break;
|
||||
case ALGO_JHA: register_jha_algo ( gate ); break;
|
||||
case ALGO_KECCAK: register_keccak_algo ( gate ); break;
|
||||
case ALGO_KECCAKC: register_keccakc_algo ( gate ); break;
|
||||
case ALGO_LBRY: register_lbry_algo ( gate ); break;
|
||||
case ALGO_LUFFA: register_luffa_algo ( gate ); break;
|
||||
case ALGO_LYRA2H: register_lyra2h_algo ( gate ); break;
|
||||
case ALGO_LYRA2RE: register_lyra2re_algo ( gate ); break;
|
||||
case ALGO_LYRA2REV2: register_lyra2rev2_algo ( gate ); break;
|
||||
case ALGO_LYRA2REV3: register_lyra2rev3_algo ( gate ); break;
|
||||
case ALGO_LYRA2Z: register_lyra2z_algo ( gate ); break;
|
||||
case ALGO_LYRA2Z330: register_lyra2z330_algo ( gate ); break;
|
||||
case ALGO_M7M: register_m7m_algo ( gate ); break;
|
||||
case ALGO_MYR_GR: register_myriad_algo ( gate ); break;
|
||||
case ALGO_NEOSCRYPT: register_neoscrypt_algo ( gate ); break;
|
||||
case ALGO_NIST5: register_nist5_algo ( gate ); break;
|
||||
case ALGO_PENTABLAKE: register_pentablake_algo ( gate ); break;
|
||||
case ALGO_PHI1612: register_phi1612_algo ( gate ); break;
|
||||
case ALGO_PHI2: register_phi2_algo ( gate ); break;
|
||||
case ALGO_PLUCK: register_pluck_algo ( gate ); break;
|
||||
case ALGO_POLYTIMOS: register_polytimos_algo ( gate ); break;
|
||||
case ALGO_QUARK: register_quark_algo ( gate ); break;
|
||||
case ALGO_QUBIT: register_qubit_algo ( gate ); break;
|
||||
case ALGO_SCRYPT: register_scrypt_algo ( gate ); break;
|
||||
case ALGO_SCRYPTJANE: register_scryptjane_algo ( gate ); break;
|
||||
case ALGO_SHA256D: register_sha256d_algo ( gate ); break;
|
||||
case ALGO_SHA256T: register_sha256t_algo ( gate ); break;
|
||||
case ALGO_SHA256Q: register_sha256q_algo ( gate ); break;
|
||||
case ALGO_SHAVITE3: register_shavite_algo ( gate ); break;
|
||||
case ALGO_SKEIN: register_skein_algo ( gate ); break;
|
||||
case ALGO_SKEIN2: register_skein2_algo ( gate ); break;
|
||||
case ALGO_SKUNK: register_skunk_algo ( gate ); break;
|
||||
case ALGO_SONOA: register_sonoa_algo ( gate ); break;
|
||||
case ALGO_TIMETRAVEL: register_timetravel_algo ( gate ); break;
|
||||
case ALGO_TIMETRAVEL10: register_timetravel10_algo ( gate ); break;
|
||||
case ALGO_TRIBUS: register_tribus_algo ( gate ); break;
|
||||
case ALGO_VANILLA: register_vanilla_algo ( gate ); break;
|
||||
case ALGO_VELTOR: register_veltor_algo ( gate ); break;
|
||||
case ALGO_WHIRLPOOL: register_whirlpool_algo ( gate ); break;
|
||||
case ALGO_WHIRLPOOLX: register_whirlpoolx_algo ( gate ); break;
|
||||
case ALGO_X11: register_x11_algo ( gate ); break;
|
||||
case ALGO_X11EVO: register_x11evo_algo ( gate ); break;
|
||||
case ALGO_X11GOST: register_x11gost_algo ( gate ); break;
|
||||
case ALGO_X12: register_x12_algo ( gate ); break;
|
||||
case ALGO_X13: register_x13_algo ( gate ); break;
|
||||
case ALGO_X13SM3: register_x13sm3_algo ( gate ); break;
|
||||
case ALGO_X14: register_x14_algo ( gate ); break;
|
||||
case ALGO_X15: register_x15_algo ( gate ); break;
|
||||
case ALGO_X16R: register_x16r_algo ( gate ); break;
|
||||
case ALGO_X16S: register_x16s_algo ( gate ); break;
|
||||
case ALGO_X17: register_x17_algo ( gate ); break;
|
||||
case ALGO_XEVAN: register_xevan_algo ( gate ); break;
|
||||
/* case ALGO_YESCRYPT: register_yescrypt_05_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR8: register_yescryptr8_05_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR16: register_yescryptr16_05_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR32: register_yescryptr32_05_algo ( gate ); break;
|
||||
*/
|
||||
case ALGO_YESCRYPT: register_yescrypt_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR8: register_yescryptr8_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR16: register_yescryptr16_algo ( gate ); break;
|
||||
case ALGO_YESCRYPTR32: register_yescryptr32_algo ( gate ); break;
|
||||
|
||||
case ALGO_YESPOWER: register_yespower_algo ( gate ); break;
|
||||
case ALGO_YESPOWERR16: register_yespowerr16_algo ( gate ); break;
|
||||
case ALGO_ZR5: register_zr5_algo ( gate ); break;
|
||||
default:
|
||||
applog(LOG_ERR,"FAIL: algo_gate registration failed, unknown algo %s.\n", algo_names[opt_algo] );
|
||||
return false;
|
||||
@@ -249,6 +268,10 @@ bool register_algo_gate( int algo, algo_gate_t *gate )
|
||||
// override std defaults with jr2 defaults
|
||||
bool register_json_rpc2( algo_gate_t *gate )
|
||||
{
|
||||
applog(LOG_WARNING,"\nCryptonight algorithm and variants are no longer");
|
||||
applog(LOG_WARNING,"supported by cpuminer-opt. Shares submitted will");
|
||||
applog(LOG_WARNING,"likely be rejected. Proceed at your own risk.\n");
|
||||
|
||||
gate->wait_for_diff = (void*)&do_nothing;
|
||||
gate->get_new_work = (void*)&jr2_get_new_work;
|
||||
gate->get_nonceptr = (void*)&jr2_get_nonceptr;
|
||||
@@ -285,6 +308,9 @@ void exec_hash_function( int algo, void *output, const void *pdata )
|
||||
const char* const algo_alias_map[][2] =
|
||||
{
|
||||
// alias proper
|
||||
{ "argon2d-crds", "argon2d250" },
|
||||
{ "argon2d-dyn", "argon2d500" },
|
||||
{ "argon2d-uis", "argon2d4096" },
|
||||
{ "bitcore", "timetravel10" },
|
||||
{ "bitzeny", "yescryptr8" },
|
||||
{ "blake256r8", "blakecoin" },
|
||||
@@ -302,6 +328,7 @@ const char* const algo_alias_map[][2] =
|
||||
{ "jane", "scryptjane" },
|
||||
{ "lyra2", "lyra2re" },
|
||||
{ "lyra2v2", "lyra2rev2" },
|
||||
{ "lyra2v3", "lyra2rev3" },
|
||||
{ "lyra2zoin", "lyra2z330" },
|
||||
{ "myrgr", "myr-gr" },
|
||||
{ "myriad", "myr-gr" },
|
||||
@@ -318,9 +345,9 @@ const char* const algo_alias_map[][2] =
|
||||
{ NULL, NULL }
|
||||
};
|
||||
|
||||
// if arg is a valid alias for a known algo it is updated with the proper name.
|
||||
// No validation of the algo or alias is done, It is the responsinility of the
|
||||
// calling function to validate the algo after return.
|
||||
// if arg is a valid alias for a known algo it is updated with the proper
|
||||
// name. No validation of the algo or alias is done, It is the responsinility
|
||||
// of the calling function to validate the algo after return.
|
||||
void get_algo_alias( char** algo_or_alias )
|
||||
{
|
||||
int i;
|
||||
@@ -333,3 +360,24 @@ void get_algo_alias( char** algo_or_alias )
|
||||
}
|
||||
}
|
||||
|
||||
#undef ALIAS
|
||||
#undef PROPER
|
||||
|
||||
// only for parallel when there are lanes.
|
||||
bool submit_solution( struct work *work, void *hash,
|
||||
struct thr_info *thr, int lane )
|
||||
{
|
||||
work_set_target_ratio( work, hash );
|
||||
if ( submit_work( thr, work ) )
|
||||
{
|
||||
applog( LOG_NOTICE, "Share %d submitted by thread %d, lane %d.",
|
||||
accepted_share_count + rejected_share_count + 1,
|
||||
thr->id, lane );
|
||||
return true;
|
||||
}
|
||||
else
|
||||
applog( LOG_WARNING, "Failed to submit share." );
|
||||
return false;
|
||||
}
|
||||
|
||||
|
||||
|
||||
@@ -2,6 +2,7 @@
|
||||
#include <stdbool.h>
|
||||
#include <stdint.h>
|
||||
#include "miner.h"
|
||||
#include "simd-utils.h"
|
||||
|
||||
/////////////////////////////
|
||||
////
|
||||
@@ -87,10 +88,11 @@ typedef uint32_t set_t;
|
||||
#define EMPTY_SET 0
|
||||
#define SSE2_OPT 1
|
||||
#define AES_OPT 2
|
||||
#define AVX_OPT 4
|
||||
#define AVX2_OPT 8
|
||||
#define SHA_OPT 0x10
|
||||
//#define FOUR_WAY_OPT 0x20
|
||||
#define SSE42_OPT 4
|
||||
#define AVX_OPT 8
|
||||
#define AVX2_OPT 0x10
|
||||
#define SHA_OPT 0x20
|
||||
#define AVX512_OPT 0x40
|
||||
|
||||
// return set containing all elements from sets a & b
|
||||
inline set_t set_union ( set_t a, set_t b ) { return a | b; }
|
||||
@@ -106,8 +108,15 @@ inline bool set_excl ( set_t a, set_t b ) { return (a & b) == 0; }
|
||||
|
||||
typedef struct
|
||||
{
|
||||
// special case, only one target, provides a callback for scanhash to
|
||||
// submit work with less overhead.
|
||||
// bool (*submit_work ) ( struct thr_info*, const struct work* );
|
||||
|
||||
// mandatory functions, must be overwritten
|
||||
int ( *scanhash ) ( int, struct work*, uint32_t, uint64_t* );
|
||||
// Added a 5th arg for the thread_info structure to replace the int thr id
|
||||
// in the first arg. Both will co-exist during the trasition.
|
||||
//int ( *scanhash ) ( int, struct work*, uint32_t, uint64_t* );
|
||||
int ( *scanhash ) ( int, struct work*, uint32_t, uint64_t*, struct thr_info* );
|
||||
|
||||
// optional unsafe, must be overwritten if algo uses function
|
||||
void ( *hash ) ( void*, const void*, uint32_t ) ;
|
||||
@@ -119,7 +128,7 @@ void ( *stratum_gen_work ) ( struct stratum_ctx*, struct work* );
|
||||
void ( *get_new_work ) ( struct work*, struct work*, int, uint32_t*,
|
||||
bool );
|
||||
uint32_t *( *get_nonceptr ) ( uint32_t* );
|
||||
void ( *display_extra_data ) ( struct work*, uint64_t* );
|
||||
void ( *decode_extra_data ) ( struct work*, uint64_t* );
|
||||
void ( *wait_for_diff ) ( struct stratum_ctx* );
|
||||
int64_t ( *get_max64 ) ();
|
||||
bool ( *work_decode ) ( const json_t*, struct work* );
|
||||
@@ -128,7 +137,7 @@ bool ( *submit_getwork_result ) ( CURL*, struct work* );
|
||||
void ( *gen_merkle_root ) ( char*, struct stratum_ctx* );
|
||||
void ( *build_extraheader ) ( struct work*, struct stratum_ctx* );
|
||||
void ( *build_block_header ) ( struct work*, uint32_t, uint32_t*,
|
||||
uint32_t*, uint32_t, uint32_t );
|
||||
uint32_t*, uint32_t, uint32_t );
|
||||
void ( *build_stratum_request ) ( char*, struct work*, struct stratum_ctx* );
|
||||
char* ( *malloc_txs_request ) ( struct work* );
|
||||
void ( *set_work_data_endian ) ( struct work* );
|
||||
@@ -139,10 +148,10 @@ bool ( *do_this_thread ) ( int );
|
||||
json_t* (*longpoll_rpc_call) ( CURL*, int*, char* );
|
||||
bool ( *stratum_handle_response )( json_t* );
|
||||
set_t optimizations;
|
||||
int ( *get_work_data_size ) ();
|
||||
int ntime_index;
|
||||
int nbits_index;
|
||||
int nonce_index; // use with caution, see warning below
|
||||
int work_data_size;
|
||||
int work_cmp_size;
|
||||
|
||||
} algo_gate_t;
|
||||
@@ -185,6 +194,12 @@ void four_way_not_tested();
|
||||
// allways returns failure
|
||||
int null_scanhash();
|
||||
|
||||
// The one and only, a callback for scanhash.
|
||||
bool submit_solution( struct work *work, void *hash,
|
||||
struct thr_info *thr, int lane );
|
||||
|
||||
bool submit_work( struct thr_info *thr, const struct work *work_in );
|
||||
|
||||
// displays warning
|
||||
void null_hash ();
|
||||
void null_hash_suw();
|
||||
@@ -239,8 +254,8 @@ void set_work_data_big_endian( struct work *work );
|
||||
double std_calc_network_diff( struct work *work );
|
||||
|
||||
void std_build_block_header( struct work* g_work, uint32_t version,
|
||||
uint32_t *prevhash, uint32_t *merkle_root,
|
||||
uint32_t ntime, uint32_t nbits );
|
||||
uint32_t *prevhash, uint32_t *merkle_root,
|
||||
uint32_t ntime, uint32_t nbits );
|
||||
|
||||
void std_build_extraheader( struct work *work, struct stratum_ctx *sctx );
|
||||
|
||||
@@ -253,6 +268,8 @@ bool jr2_stratum_handle_response( json_t *val );
|
||||
bool std_ready_to_mine( struct work* work, struct stratum_ctx* stratum,
|
||||
int thr_id );
|
||||
|
||||
int std_get_work_data_size();
|
||||
|
||||
// Gate admin functions
|
||||
|
||||
// Called from main to initialize all gate functions and algo-specific data
|
||||
|
||||
@@ -99,18 +99,18 @@ static const char *Argon2_ErrorMessage[] = {
|
||||
{ARGON2_MISSING_ARGS, */ "Missing arguments", /*},*/
|
||||
};
|
||||
|
||||
int argon2d(argon2_context *context) { return argon2_core(context, Argon2_d); }
|
||||
int argon2d(argon2_context *context) { return ar2_argon2_core(context, Argon2_d); }
|
||||
|
||||
int argon2i(argon2_context *context) { return argon2_core(context, Argon2_i); }
|
||||
int argon2i(argon2_context *context) { return ar2_argon2_core(context, Argon2_i); }
|
||||
|
||||
int verify_d(argon2_context *context, const char *hash)
|
||||
int ar2_verify_d(argon2_context *context, const char *hash)
|
||||
{
|
||||
int result;
|
||||
/*if (0 == context->outlen || NULL == hash) {
|
||||
return ARGON2_OUT_PTR_MISMATCH;
|
||||
}*/
|
||||
|
||||
result = argon2_core(context, Argon2_d);
|
||||
result = ar2_argon2_core(context, Argon2_d);
|
||||
|
||||
if (ARGON2_OK != result) {
|
||||
return result;
|
||||
@@ -223,7 +223,7 @@ static size_t to_base64(char *dst, size_t dst_len, const void *src)
|
||||
* The output length is always exactly 32 bytes.
|
||||
*/
|
||||
|
||||
int encode_string(char *dst, size_t dst_len, argon2_context *ctx)
|
||||
int ar2_encode_string(char *dst, size_t dst_len, argon2_context *ctx)
|
||||
{
|
||||
#define SS(str) \
|
||||
do { \
|
||||
@@ -255,7 +255,7 @@ int argon2id(argon2_context *context);
|
||||
* specified by the context outlen member
|
||||
* @return Zero if successful, a non zero error code otherwise
|
||||
*/
|
||||
int verify_d(argon2_context *context, const char *hash);
|
||||
int ar2_verify_d(argon2_context *context, const char *hash);
|
||||
|
||||
/*
|
||||
* Get the associated error message for given error code
|
||||
@@ -283,7 +283,7 @@ const char *error_message(int error_code);
|
||||
* The output length is always exactly 32 bytes.
|
||||
*/
|
||||
|
||||
int encode_string(char *dst, size_t dst_len, argon2_context *ctx);
|
||||
int ar2_encode_string(char *dst, size_t dst_len, argon2_context *ctx);
|
||||
|
||||
#if defined(__cplusplus)
|
||||
}
|
||||
@@ -52,22 +52,22 @@ enum {
|
||||
};
|
||||
|
||||
/* Streaming API */
|
||||
int blake2b_init(blake2b_state *S, size_t outlen);
|
||||
int blake2b_init_key(blake2b_state *S, size_t outlen, const void *key,
|
||||
int ar2_blake2b_init(blake2b_state *S, size_t outlen);
|
||||
int ar2_blake2b_init_key(blake2b_state *S, size_t outlen, const void *key,
|
||||
size_t keylen);
|
||||
int blake2b_init_param(blake2b_state *S, const blake2b_param *P);
|
||||
int blake2b_update(blake2b_state *S, const void *in, size_t inlen);
|
||||
int ar2_blake2b_init_param(blake2b_state *S, const blake2b_param *P);
|
||||
int ar2_blake2b_update(blake2b_state *S, const void *in, size_t inlen);
|
||||
void my_blake2b_update(blake2b_state *S, const void *in, size_t inlen);
|
||||
int blake2b_final(blake2b_state *S, void *out, size_t outlen);
|
||||
int ar2_blake2b_final(blake2b_state *S, void *out, size_t outlen);
|
||||
|
||||
/* Simple API */
|
||||
int blake2b(void *out, const void *in, const void *key, size_t keylen);
|
||||
int ar2_blake2b(void *out, const void *in, const void *key, size_t keylen);
|
||||
|
||||
/* Argon2 Team - Begin Code */
|
||||
int blake2b_long(void *out, const void *in);
|
||||
int ar2_blake2b_long(void *out, const void *in);
|
||||
/* Argon2 Team - End Code */
|
||||
/* Miouyouyou */
|
||||
void blake2b_too(void *out, const void *in);
|
||||
void ar2_blake2b_too(void *out, const void *in);
|
||||
|
||||
#if defined(__cplusplus)
|
||||
}
|
||||
@@ -107,7 +107,7 @@ static const blake2b_state miou = {
|
||||
};
|
||||
|
||||
|
||||
int blake2b_init_param(blake2b_state *S, const blake2b_param *P)
|
||||
int ar2_blake2b_init_param(blake2b_state *S, const blake2b_param *P)
|
||||
{
|
||||
const unsigned char *p = (const unsigned char *)P;
|
||||
unsigned int i;
|
||||
@@ -133,7 +133,7 @@ void compare_buffs(uint64_t *h, size_t outlen)
|
||||
}
|
||||
|
||||
/* Sequential blake2b initialization */
|
||||
int blake2b_init(blake2b_state *S, size_t outlen)
|
||||
int ar2_blake2b_init(blake2b_state *S, size_t outlen)
|
||||
{
|
||||
memcpy(S, &miou, sizeof(*S));
|
||||
S->h[0] += outlen;
|
||||
@@ -147,7 +147,7 @@ void print64(const char *name, const uint64_t *array, uint16_t size)
|
||||
printf("};\n");
|
||||
}
|
||||
|
||||
int blake2b_init_key(blake2b_state *S, size_t outlen, const void *key, size_t keylen)
|
||||
int ar2_blake2b_init_key(blake2b_state *S, size_t outlen, const void *key, size_t keylen)
|
||||
{
|
||||
return 0;
|
||||
}
|
||||
@@ -207,7 +207,7 @@ static void blake2b_compress(blake2b_state *S, const uint8_t *block)
|
||||
#undef ROUND
|
||||
}
|
||||
|
||||
int blake2b_update(blake2b_state *S, const void *in, size_t inlen)
|
||||
int ar2_blake2b_update(blake2b_state *S, const void *in, size_t inlen)
|
||||
{
|
||||
const uint8_t *pin = (const uint8_t *)in;
|
||||
/* Complete current block */
|
||||
@@ -235,7 +235,7 @@ void my_blake2b_update(blake2b_state *S, const void *in, size_t inlen)
|
||||
S->buflen += (unsigned int)inlen;
|
||||
}
|
||||
|
||||
int blake2b_final(blake2b_state *S, void *out, size_t outlen)
|
||||
int ar2_blake2b_final(blake2b_state *S, void *out, size_t outlen)
|
||||
{
|
||||
uint8_t buffer[BLAKE2B_OUTBYTES] = {0};
|
||||
unsigned int i;
|
||||
@@ -257,48 +257,48 @@ int blake2b_final(blake2b_state *S, void *out, size_t outlen)
|
||||
return 0;
|
||||
}
|
||||
|
||||
int blake2b(void *out, const void *in, const void *key, size_t keylen)
|
||||
int ar2_blake2b(void *out, const void *in, const void *key, size_t keylen)
|
||||
{
|
||||
blake2b_state S;
|
||||
|
||||
blake2b_init(&S, 64);
|
||||
ar2_blake2b_init(&S, 64);
|
||||
my_blake2b_update(&S, in, 64);
|
||||
blake2b_final(&S, out, 64);
|
||||
ar2_blake2b_final(&S, out, 64);
|
||||
burn(&S, sizeof(S));
|
||||
return 0;
|
||||
}
|
||||
|
||||
void blake2b_too(void *pout, const void *in)
|
||||
void ar2_blake2b_too(void *pout, const void *in)
|
||||
{
|
||||
uint8_t *out = (uint8_t *)pout;
|
||||
uint8_t out_buffer[64];
|
||||
uint8_t in_buffer[64];
|
||||
|
||||
blake2b_state blake_state;
|
||||
blake2b_init(&blake_state, 64);
|
||||
ar2_blake2b_init(&blake_state, 64);
|
||||
blake_state.buflen = blake_state.buf[1] = 4;
|
||||
my_blake2b_update(&blake_state, in, 72);
|
||||
blake2b_final(&blake_state, out_buffer, 64);
|
||||
ar2_blake2b_final(&blake_state, out_buffer, 64);
|
||||
memcpy(out, out_buffer, 32);
|
||||
out += 32;
|
||||
|
||||
register uint8_t i = 29;
|
||||
while (i--) {
|
||||
memcpy(in_buffer, out_buffer, 64);
|
||||
blake2b(out_buffer, in_buffer, NULL, 0);
|
||||
ar2_blake2b(out_buffer, in_buffer, NULL, 0);
|
||||
memcpy(out, out_buffer, 32);
|
||||
out += 32;
|
||||
}
|
||||
|
||||
memcpy(in_buffer, out_buffer, 64);
|
||||
blake2b(out_buffer, in_buffer, NULL, 0);
|
||||
ar2_blake2b(out_buffer, in_buffer, NULL, 0);
|
||||
memcpy(out, out_buffer, 64);
|
||||
|
||||
burn(&blake_state, sizeof(blake_state));
|
||||
}
|
||||
|
||||
/* Argon2 Team - Begin Code */
|
||||
int blake2b_long(void *pout, const void *in)
|
||||
int ar2_blake2b_long(void *pout, const void *in)
|
||||
{
|
||||
uint8_t *out = (uint8_t *)pout;
|
||||
blake2b_state blake_state;
|
||||
@@ -306,10 +306,10 @@ int blake2b_long(void *pout, const void *in)
|
||||
|
||||
store32(outlen_bytes, 32);
|
||||
|
||||
blake2b_init(&blake_state, 32);
|
||||
ar2_blake2b_init(&blake_state, 32);
|
||||
my_blake2b_update(&blake_state, outlen_bytes, sizeof(outlen_bytes));
|
||||
blake2b_update(&blake_state, in, 1024);
|
||||
blake2b_final(&blake_state, out, 32);
|
||||
ar2_blake2b_update(&blake_state, in, 1024);
|
||||
ar2_blake2b_final(&blake_state, out, 32);
|
||||
burn(&blake_state, sizeof(blake_state));
|
||||
return 0;
|
||||
}
|
||||
@@ -51,15 +51,15 @@
|
||||
#endif
|
||||
|
||||
/***************Instance and Position constructors**********/
|
||||
void init_block_value(block *b, uint8_t in) { memset(b->v, in, sizeof(b->v)); }
|
||||
void ar2_init_block_value(block *b, uint8_t in) { memset(b->v, in, sizeof(b->v)); }
|
||||
//inline void init_block_value(block *b, uint8_t in) { memset(b->v, in, sizeof(b->v)); }
|
||||
|
||||
void copy_block(block *dst, const block *src) {
|
||||
void ar2_copy_block(block *dst, const block *src) {
|
||||
//inline void copy_block(block *dst, const block *src) {
|
||||
memcpy(dst->v, src->v, sizeof(uint64_t) * ARGON2_WORDS_IN_BLOCK);
|
||||
}
|
||||
|
||||
void xor_block(block *dst, const block *src) {
|
||||
void ar2_xor_block(block *dst, const block *src) {
|
||||
//inline void xor_block(block *dst, const block *src) {
|
||||
int i;
|
||||
for (i = 0; i < ARGON2_WORDS_IN_BLOCK; ++i) {
|
||||
@@ -67,7 +67,7 @@ void xor_block(block *dst, const block *src) {
|
||||
}
|
||||
}
|
||||
|
||||
static void load_block(block *dst, const void *input) {
|
||||
static void ar2_load_block(block *dst, const void *input) {
|
||||
//static inline void load_block(block *dst, const void *input) {
|
||||
unsigned i;
|
||||
for (i = 0; i < ARGON2_WORDS_IN_BLOCK; ++i) {
|
||||
@@ -75,7 +75,7 @@ static void load_block(block *dst, const void *input) {
|
||||
}
|
||||
}
|
||||
|
||||
static void store_block(void *output, const block *src) {
|
||||
static void ar2_store_block(void *output, const block *src) {
|
||||
//static inline void store_block(void *output, const block *src) {
|
||||
unsigned i;
|
||||
for (i = 0; i < ARGON2_WORDS_IN_BLOCK; ++i) {
|
||||
@@ -84,7 +84,7 @@ static void store_block(void *output, const block *src) {
|
||||
}
|
||||
|
||||
/***************Memory allocators*****************/
|
||||
int allocate_memory(block **memory, uint32_t m_cost) {
|
||||
int ar2_allocate_memory(block **memory, uint32_t m_cost) {
|
||||
if (memory != NULL) {
|
||||
size_t memory_size = sizeof(block) * m_cost;
|
||||
if (m_cost != 0 &&
|
||||
@@ -105,34 +105,34 @@ int allocate_memory(block **memory, uint32_t m_cost) {
|
||||
}
|
||||
}
|
||||
|
||||
void secure_wipe_memory(void *v, size_t n) { memset(v, 0, n); }
|
||||
void ar2_secure_wipe_memory(void *v, size_t n) { memset(v, 0, n); }
|
||||
//inline void secure_wipe_memory(void *v, size_t n) { memset(v, 0, n); }
|
||||
|
||||
/*********Memory functions*/
|
||||
|
||||
void clear_memory(argon2_instance_t *instance, int clear) {
|
||||
void ar2_clear_memory(argon2_instance_t *instance, int clear) {
|
||||
//inline void clear_memory(argon2_instance_t *instance, int clear) {
|
||||
if (instance->memory != NULL && clear) {
|
||||
secure_wipe_memory(instance->memory,
|
||||
ar2_secure_wipe_memory(instance->memory,
|
||||
sizeof(block) * /*instance->memory_blocks*/16);
|
||||
}
|
||||
}
|
||||
|
||||
void free_memory(block *memory) { free(memory); }
|
||||
void ar2_free_memory(block *memory) { free(memory); }
|
||||
//inline void free_memory(block *memory) { free(memory); }
|
||||
|
||||
void finalize(const argon2_context *context, argon2_instance_t *instance) {
|
||||
void ar2_finalize(const argon2_context *context, argon2_instance_t *instance) {
|
||||
if (context != NULL && instance != NULL) {
|
||||
block blockhash;
|
||||
copy_block(&blockhash, instance->memory + 15);
|
||||
ar2_copy_block(&blockhash, instance->memory + 15);
|
||||
|
||||
/* Hash the result */
|
||||
{
|
||||
uint8_t blockhash_bytes[ARGON2_BLOCK_SIZE];
|
||||
store_block(blockhash_bytes, &blockhash);
|
||||
blake2b_long(context->out, blockhash_bytes);
|
||||
secure_wipe_memory(blockhash.v, ARGON2_BLOCK_SIZE);
|
||||
secure_wipe_memory(blockhash_bytes, ARGON2_BLOCK_SIZE); /* clear blockhash_bytes */
|
||||
ar2_store_block(blockhash_bytes, &blockhash);
|
||||
ar2_blake2b_long(context->out, blockhash_bytes);
|
||||
ar2_secure_wipe_memory(blockhash.v, ARGON2_BLOCK_SIZE);
|
||||
ar2_secure_wipe_memory(blockhash_bytes, ARGON2_BLOCK_SIZE); /* clear blockhash_bytes */
|
||||
}
|
||||
|
||||
#ifdef GENKAT
|
||||
@@ -142,11 +142,11 @@ void finalize(const argon2_context *context, argon2_instance_t *instance) {
|
||||
/* Clear memory */
|
||||
// clear_memory(instance, 1);
|
||||
|
||||
free_memory(instance->memory);
|
||||
ar2_free_memory(instance->memory);
|
||||
}
|
||||
}
|
||||
|
||||
uint32_t index_alpha(const argon2_instance_t *instance,
|
||||
uint32_t ar2_index_alpha(const argon2_instance_t *instance,
|
||||
const argon2_position_t *position, uint32_t pseudo_rand,
|
||||
int same_lane) {
|
||||
/*
|
||||
@@ -207,7 +207,7 @@ uint32_t index_alpha(const argon2_instance_t *instance,
|
||||
return absolute_position;
|
||||
}
|
||||
|
||||
void fill_memory_blocks(argon2_instance_t *instance) {
|
||||
void ar2_fill_memory_blocks(argon2_instance_t *instance) {
|
||||
uint32_t r, s;
|
||||
|
||||
for (r = 0; r < 2; ++r) {
|
||||
@@ -218,7 +218,7 @@ void fill_memory_blocks(argon2_instance_t *instance) {
|
||||
position.lane = 0;
|
||||
position.slice = (uint8_t)s;
|
||||
position.index = 0;
|
||||
fill_segment(instance, position);
|
||||
ar2_fill_segment(instance, position);
|
||||
}
|
||||
|
||||
#ifdef GENKAT
|
||||
@@ -227,19 +227,19 @@ void fill_memory_blocks(argon2_instance_t *instance) {
|
||||
}
|
||||
}
|
||||
|
||||
void fill_first_blocks(uint8_t *blockhash, const argon2_instance_t *instance) {
|
||||
void ar2_fill_first_blocks(uint8_t *blockhash, const argon2_instance_t *instance) {
|
||||
/* Make the first and second block in each lane as G(H0||i||0) or
|
||||
G(H0||i||1) */
|
||||
uint8_t blockhash_bytes[ARGON2_BLOCK_SIZE];
|
||||
store32(blockhash + ARGON2_PREHASH_DIGEST_LENGTH, 0);
|
||||
store32(blockhash + ARGON2_PREHASH_DIGEST_LENGTH + 4, 0);
|
||||
blake2b_too(blockhash_bytes, blockhash);
|
||||
load_block(&instance->memory[0], blockhash_bytes);
|
||||
ar2_blake2b_too(blockhash_bytes, blockhash);
|
||||
ar2_load_block(&instance->memory[0], blockhash_bytes);
|
||||
|
||||
store32(blockhash + ARGON2_PREHASH_DIGEST_LENGTH, 1);
|
||||
blake2b_too(blockhash_bytes, blockhash);
|
||||
load_block(&instance->memory[1], blockhash_bytes);
|
||||
secure_wipe_memory(blockhash_bytes, ARGON2_BLOCK_SIZE);
|
||||
ar2_blake2b_too(blockhash_bytes, blockhash);
|
||||
ar2_load_block(&instance->memory[1], blockhash_bytes);
|
||||
ar2_secure_wipe_memory(blockhash_bytes, ARGON2_BLOCK_SIZE);
|
||||
}
|
||||
|
||||
|
||||
@@ -268,7 +268,7 @@ static const blake2b_state base_hash = {
|
||||
#define SALTLEN 32
|
||||
#define SECRETLEN 0
|
||||
#define ADLEN 0
|
||||
void initial_hash(uint8_t *blockhash, argon2_context *context,
|
||||
void ar2_initial_hash(uint8_t *blockhash, argon2_context *context,
|
||||
argon2_type type) {
|
||||
|
||||
uint8_t value[sizeof(uint32_t)];
|
||||
@@ -280,7 +280,7 @@ void initial_hash(uint8_t *blockhash, argon2_context *context,
|
||||
PWDLEN);
|
||||
|
||||
|
||||
secure_wipe_memory(context->pwd, PWDLEN);
|
||||
ar2_secure_wipe_memory(context->pwd, PWDLEN);
|
||||
context->pwdlen = 0;
|
||||
|
||||
store32(&value, SALTLEN);
|
||||
@@ -295,22 +295,22 @@ void initial_hash(uint8_t *blockhash, argon2_context *context,
|
||||
store32(&value, ADLEN);
|
||||
my_blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
blake2b_final(&BlakeHash, blockhash, ARGON2_PREHASH_DIGEST_LENGTH);
|
||||
ar2_blake2b_final(&BlakeHash, blockhash, ARGON2_PREHASH_DIGEST_LENGTH);
|
||||
}
|
||||
|
||||
int initialize(argon2_instance_t *instance, argon2_context *context) {
|
||||
int ar2_initialize(argon2_instance_t *instance, argon2_context *context) {
|
||||
/* 1. Memory allocation */
|
||||
|
||||
|
||||
allocate_memory(&(instance->memory), 16);
|
||||
ar2_allocate_memory(&(instance->memory), 16);
|
||||
|
||||
/* 2. Initial hashing */
|
||||
/* H_0 + 8 extra bytes to produce the first blocks */
|
||||
/* Hashing all inputs */
|
||||
uint8_t blockhash[ARGON2_PREHASH_SEED_LENGTH];
|
||||
initial_hash(blockhash, context, instance->type);
|
||||
ar2_initial_hash(blockhash, context, instance->type);
|
||||
/* Zeroing 8 extra bytes */
|
||||
secure_wipe_memory(blockhash + ARGON2_PREHASH_DIGEST_LENGTH,
|
||||
ar2_secure_wipe_memory(blockhash + ARGON2_PREHASH_DIGEST_LENGTH,
|
||||
ARGON2_PREHASH_SEED_LENGTH -
|
||||
ARGON2_PREHASH_DIGEST_LENGTH);
|
||||
|
||||
@@ -320,14 +320,14 @@ int initialize(argon2_instance_t *instance, argon2_context *context) {
|
||||
|
||||
/* 3. Creating first blocks, we always have at least two blocks in a slice
|
||||
*/
|
||||
fill_first_blocks(blockhash, instance);
|
||||
ar2_fill_first_blocks(blockhash, instance);
|
||||
/* Clearing the hash */
|
||||
secure_wipe_memory(blockhash, ARGON2_PREHASH_SEED_LENGTH);
|
||||
ar2_secure_wipe_memory(blockhash, ARGON2_PREHASH_SEED_LENGTH);
|
||||
|
||||
return ARGON2_OK;
|
||||
}
|
||||
|
||||
int argon2_core(argon2_context *context, argon2_type type) {
|
||||
int ar2_argon2_core(argon2_context *context, argon2_type type) {
|
||||
argon2_instance_t instance;
|
||||
instance.memory = NULL;
|
||||
instance.type = type;
|
||||
@@ -336,14 +336,14 @@ int argon2_core(argon2_context *context, argon2_type type) {
|
||||
* blocks
|
||||
*/
|
||||
|
||||
int result = initialize(&instance, context);
|
||||
int result = ar2_initialize(&instance, context);
|
||||
if (ARGON2_OK != result) return result;
|
||||
|
||||
/* 4. Filling memory */
|
||||
fill_memory_blocks(&instance);
|
||||
ar2_fill_memory_blocks(&instance);
|
||||
|
||||
/* 5. Finalization */
|
||||
finalize(context, &instance);
|
||||
ar2_finalize(context, &instance);
|
||||
|
||||
return ARGON2_OK;
|
||||
}
|
||||
@@ -62,13 +62,13 @@ typedef struct _block { uint64_t v[ARGON2_WORDS_IN_BLOCK]; } ALIGN(16) block;
|
||||
/*****************Functions that work with the block******************/
|
||||
|
||||
/* Initialize each byte of the block with @in */
|
||||
void init_block_value(block *b, uint8_t in);
|
||||
void ar2_init_block_value(block *b, uint8_t in);
|
||||
|
||||
/* Copy block @src to block @dst */
|
||||
void copy_block(block *dst, const block *src);
|
||||
void ar2_copy_block(block *dst, const block *src);
|
||||
|
||||
/* XOR @src onto @dst bytewise */
|
||||
void xor_block(block *dst, const block *src);
|
||||
void ar2_xor_block(block *dst, const block *src);
|
||||
|
||||
/*
|
||||
* Argon2 instance: memory pointer, number of passes, amount of memory, type,
|
||||
@@ -101,24 +101,24 @@ typedef struct Argon2_position_t {
|
||||
* @param m_cost number of blocks to allocate in the memory
|
||||
* @return ARGON2_OK if @memory is a valid pointer and memory is allocated
|
||||
*/
|
||||
int allocate_memory(block **memory, uint32_t m_cost);
|
||||
int ar2_allocate_memory(block **memory, uint32_t m_cost);
|
||||
|
||||
/* Function that securely cleans the memory
|
||||
* @param mem Pointer to the memory
|
||||
* @param s Memory size in bytes
|
||||
*/
|
||||
void secure_wipe_memory(void *v, size_t n);
|
||||
void ar2_secure_wipe_memory(void *v, size_t n);
|
||||
|
||||
/* Clears memory
|
||||
* @param instance pointer to the current instance
|
||||
* @param clear_memory indicates if we clear the memory with zeros.
|
||||
*/
|
||||
void clear_memory(argon2_instance_t *instance, int clear);
|
||||
void ar2_clear_memory(argon2_instance_t *instance, int clear);
|
||||
|
||||
/* Deallocates memory
|
||||
* @param memory pointer to the blocks
|
||||
*/
|
||||
void free_memory(block *memory);
|
||||
void ar2_free_memory(block *memory);
|
||||
|
||||
/*
|
||||
* Computes absolute position of reference block in the lane following a skewed
|
||||
@@ -130,7 +130,7 @@ void free_memory(block *memory);
|
||||
* If so we can reference the current segment
|
||||
* @pre All pointers must be valid
|
||||
*/
|
||||
uint32_t index_alpha(const argon2_instance_t *instance,
|
||||
uint32_t ar2_index_alpha(const argon2_instance_t *instance,
|
||||
const argon2_position_t *position, uint32_t pseudo_rand,
|
||||
int same_lane);
|
||||
|
||||
@@ -141,7 +141,7 @@ uint32_t index_alpha(const argon2_instance_t *instance,
|
||||
* @return ARGON2_OK if everything is all right, otherwise one of error codes
|
||||
* (all defined in <argon2.h>
|
||||
*/
|
||||
int validate_inputs(const argon2_context *context);
|
||||
int ar2_validate_inputs(const argon2_context *context);
|
||||
|
||||
/*
|
||||
* Hashes all the inputs into @a blockhash[PREHASH_DIGEST_LENGTH], clears
|
||||
@@ -153,7 +153,7 @@ int validate_inputs(const argon2_context *context);
|
||||
* @pre @a blockhash must have at least @a PREHASH_DIGEST_LENGTH bytes
|
||||
* allocated
|
||||
*/
|
||||
void initial_hash(uint8_t *blockhash, argon2_context *context,
|
||||
void ar2_initial_hash(uint8_t *blockhash, argon2_context *context,
|
||||
argon2_type type);
|
||||
|
||||
/*
|
||||
@@ -162,7 +162,7 @@ void initial_hash(uint8_t *blockhash, argon2_context *context,
|
||||
* @param blockhash Pointer to the pre-hashing digest
|
||||
* @pre blockhash must point to @a PREHASH_SEED_LENGTH allocated values
|
||||
*/
|
||||
void fill_firsts_blocks(uint8_t *blockhash, const argon2_instance_t *instance);
|
||||
void ar2_fill_firsts_blocks(uint8_t *blockhash, const argon2_instance_t *instance);
|
||||
|
||||
/*
|
||||
* Function allocates memory, hashes the inputs with Blake, and creates first
|
||||
@@ -174,7 +174,7 @@ void fill_firsts_blocks(uint8_t *blockhash, const argon2_instance_t *instance);
|
||||
* @return Zero if successful, -1 if memory failed to allocate. @context->state
|
||||
* will be modified if successful.
|
||||
*/
|
||||
int initialize(argon2_instance_t *instance, argon2_context *context);
|
||||
int ar2_initialize(argon2_instance_t *instance, argon2_context *context);
|
||||
|
||||
/*
|
||||
* XORing the last block of each lane, hashing it, making the tag. Deallocates
|
||||
@@ -187,7 +187,7 @@ int initialize(argon2_instance_t *instance, argon2_context *context);
|
||||
* @pre if context->free_cbk is not NULL, it should point to a function that
|
||||
* deallocates memory
|
||||
*/
|
||||
void finalize(const argon2_context *context, argon2_instance_t *instance);
|
||||
void ar2_finalize(const argon2_context *context, argon2_instance_t *instance);
|
||||
|
||||
/*
|
||||
* Function that fills the segment using previous segments also from other
|
||||
@@ -196,7 +196,7 @@ void finalize(const argon2_context *context, argon2_instance_t *instance);
|
||||
* @param position Current position
|
||||
* @pre all block pointers must be valid
|
||||
*/
|
||||
void fill_segment(const argon2_instance_t *instance,
|
||||
void ar2_fill_segment(const argon2_instance_t *instance,
|
||||
argon2_position_t position);
|
||||
|
||||
/*
|
||||
@@ -204,13 +204,13 @@ void fill_segment(const argon2_instance_t *instance,
|
||||
* blocks in each lane
|
||||
* @param instance Pointer to the current instance
|
||||
*/
|
||||
void fill_memory_blocks(argon2_instance_t *instance);
|
||||
void ar2_fill_memory_blocks(argon2_instance_t *instance);
|
||||
|
||||
/*
|
||||
* Function that performs memory-hard hashing with certain degree of parallelism
|
||||
* @param context Pointer to the Argon2 internal structure
|
||||
* @return Error code if smth is wrong, ARGON2_OK otherwise
|
||||
*/
|
||||
int argon2_core(argon2_context *context, argon2_type type);
|
||||
int ar2_argon2_core(argon2_context *context, argon2_type type);
|
||||
|
||||
#endif
|
||||
@@ -26,7 +26,7 @@
|
||||
#include "blake2/blake2.h"
|
||||
#include "blake2/blamka-round-opt.h"
|
||||
|
||||
void fill_block(__m128i *state, __m128i const *ref_block, __m128i *next_block)
|
||||
void ar2_fill_block(__m128i *state, __m128i const *ref_block, __m128i *next_block)
|
||||
{
|
||||
__m128i ALIGN(16) block_XY[ARGON2_QWORDS_IN_BLOCK];
|
||||
uint32_t i;
|
||||
@@ -95,7 +95,7 @@ static const uint64_t bad_rands[32] = {
|
||||
UINT64_C(8548260058287621283), UINT64_C(8641748798041936364)
|
||||
};
|
||||
|
||||
void generate_addresses(const argon2_instance_t *instance,
|
||||
void ar2_generate_addresses(const argon2_instance_t *instance,
|
||||
const argon2_position_t *position,
|
||||
uint64_t *pseudo_rands)
|
||||
{
|
||||
@@ -113,7 +113,7 @@ void generate_addresses(const argon2_instance_t *instance,
|
||||
#define LANE_LENGTH 16
|
||||
#define POS_LANE 0
|
||||
|
||||
void fill_segment(const argon2_instance_t *instance,
|
||||
void ar2_fill_segment(const argon2_instance_t *instance,
|
||||
argon2_position_t position)
|
||||
{
|
||||
block *ref_block = NULL, *curr_block = NULL;
|
||||
@@ -129,7 +129,7 @@ void fill_segment(const argon2_instance_t *instance,
|
||||
pseudo_rands = (uint64_t *)malloc(/*sizeof(uint64_t) * 4*/32);
|
||||
|
||||
if (data_independent_addressing) {
|
||||
generate_addresses(instance, &position, pseudo_rands);
|
||||
ar2_generate_addresses(instance, &position, pseudo_rands);
|
||||
}
|
||||
|
||||
i = 0;
|
||||
@@ -173,12 +173,12 @@ void fill_segment(const argon2_instance_t *instance,
|
||||
* lane.
|
||||
*/
|
||||
position.index = i;
|
||||
ref_index = index_alpha(instance, &position, pseudo_rand & 0xFFFFFFFF,1);
|
||||
ref_index = ar2_index_alpha(instance, &position, pseudo_rand & 0xFFFFFFFF,1);
|
||||
|
||||
/* 2 Creating a new block */
|
||||
ref_block = instance->memory + ref_index;
|
||||
curr_block = instance->memory + curr_offset;
|
||||
fill_block(state, (__m128i const *)ref_block->v, (__m128i *)curr_block->v);
|
||||
ar2_fill_block(state, (__m128i const *)ref_block->v, (__m128i *)curr_block->v);
|
||||
}
|
||||
|
||||
free(pseudo_rands);
|
||||
@@ -21,7 +21,7 @@
|
||||
* @param next_block Pointer to the block to be constructed
|
||||
* @pre all block pointers must be valid
|
||||
*/
|
||||
void fill_block(__m128i *state, __m128i const *ref_block, __m128i *next_block);
|
||||
void ar2_fill_block(__m128i *state, __m128i const *ref_block, __m128i *next_block);
|
||||
|
||||
/*
|
||||
* Generate pseudo-random values to reference blocks in the segment and puts
|
||||
@@ -31,7 +31,7 @@ void fill_block(__m128i *state, __m128i const *ref_block, __m128i *next_block);
|
||||
* @param pseudo_rands Pointer to the array of 64-bit values
|
||||
* @pre pseudo_rands must point to @a instance->segment_length allocated values
|
||||
*/
|
||||
void generate_addresses(const argon2_instance_t *instance,
|
||||
void ar2_generate_addresses(const argon2_instance_t *instance,
|
||||
const argon2_position_t *position,
|
||||
uint64_t *pseudo_rands);
|
||||
|
||||
@@ -43,7 +43,7 @@ void generate_addresses(const argon2_instance_t *instance,
|
||||
* @param position Current position
|
||||
* @pre all block pointers must be valid
|
||||
*/
|
||||
void fill_segment(const argon2_instance_t *instance,
|
||||
void ar2_fill_segment(const argon2_instance_t *instance,
|
||||
argon2_position_t position);
|
||||
|
||||
#endif /* ARGON2_OPT_H */
|
||||
@@ -24,7 +24,7 @@ inline void argon_call(void *out, void *in, void *salt, int type)
|
||||
context.allocate_cbk = NULL;
|
||||
context.free_cbk = NULL;
|
||||
|
||||
argon2_core(&context, type);
|
||||
ar2_argon2_core(&context, type);
|
||||
}
|
||||
|
||||
void argon2hash(void *output, const void *input)
|
||||
@@ -79,7 +79,7 @@ int64_t argon2_get_max64 ()
|
||||
|
||||
bool register_argon2_algo( algo_gate_t* gate )
|
||||
{
|
||||
gate->optimizations = SSE2_OPT | AES_OPT | AVX_OPT | AVX2_OPT;
|
||||
gate->optimizations = SSE2_OPT | AVX_OPT | AVX2_OPT;
|
||||
gate->scanhash = (void*)&scanhash_argon2;
|
||||
gate->hash = (void*)&argon2hash;
|
||||
gate->gen_merkle_root = (void*)&SHA256_gen_merkle_root;
|
||||
198
algo/argon2/argon2d/argon2d-gate.c
Normal file
198
algo/argon2/argon2d/argon2d-gate.c
Normal file
@@ -0,0 +1,198 @@
|
||||
#include "argon2d-gate.h"
|
||||
#include "argon2d/argon2.h"
|
||||
|
||||
static const size_t INPUT_BYTES = 80; // Lenth of a block header in bytes. Input Length = Salt Length (salt = input)
|
||||
static const size_t OUTPUT_BYTES = 32; // Length of output needed for a 256-bit hash
|
||||
static const unsigned int DEFAULT_ARGON2_FLAG = 2; //Same as ARGON2_DEFAULT_FLAGS
|
||||
|
||||
// Credits
|
||||
|
||||
void argon2d_crds_hash( void *output, const void *input )
|
||||
{
|
||||
argon2_context context;
|
||||
context.out = (uint8_t *)output;
|
||||
context.outlen = (uint32_t)OUTPUT_BYTES;
|
||||
context.pwd = (uint8_t *)input;
|
||||
context.pwdlen = (uint32_t)INPUT_BYTES;
|
||||
context.salt = (uint8_t *)input; //salt = input
|
||||
context.saltlen = (uint32_t)INPUT_BYTES;
|
||||
context.secret = NULL;
|
||||
context.secretlen = 0;
|
||||
context.ad = NULL;
|
||||
context.adlen = 0;
|
||||
context.allocate_cbk = NULL;
|
||||
context.free_cbk = NULL;
|
||||
context.flags = DEFAULT_ARGON2_FLAG; // = ARGON2_DEFAULT_FLAGS
|
||||
// main configurable Argon2 hash parameters
|
||||
context.m_cost = 250; // Memory in KiB (~256KB)
|
||||
context.lanes = 4; // Degree of Parallelism
|
||||
context.threads = 1; // Threads
|
||||
context.t_cost = 1; // Iterations
|
||||
context.version = ARGON2_VERSION_10;
|
||||
|
||||
argon2_ctx( &context, Argon2_d );
|
||||
}
|
||||
|
||||
int scanhash_argon2d_crds( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done )
|
||||
{
|
||||
uint32_t _ALIGN(64) endiandata[20];
|
||||
uint32_t _ALIGN(64) hash[8];
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
|
||||
uint32_t nonce = first_nonce;
|
||||
|
||||
swab32_array( endiandata, pdata, 20 );
|
||||
|
||||
do {
|
||||
be32enc(&endiandata[19], nonce);
|
||||
argon2d_crds_hash( hash, endiandata );
|
||||
if ( hash[7] <= Htarg && fulltest( hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = nonce;
|
||||
*hashes_done = pdata[19] - first_nonce;
|
||||
work_set_target_ratio(work, hash);
|
||||
return 1;
|
||||
}
|
||||
nonce++;
|
||||
} while (nonce < max_nonce && !work_restart[thr_id].restart);
|
||||
|
||||
pdata[19] = nonce;
|
||||
*hashes_done = pdata[19] - first_nonce + 1;
|
||||
return 0;
|
||||
}
|
||||
|
||||
bool register_argon2d_crds_algo( algo_gate_t* gate )
|
||||
{
|
||||
gate->scanhash = (void*)&scanhash_argon2d_crds;
|
||||
gate->hash = (void*)&argon2d_crds_hash;
|
||||
gate->set_target = (void*)&scrypt_set_target;
|
||||
gate->optimizations = SSE2_OPT | AVX2_OPT | AVX512_OPT;
|
||||
return true;
|
||||
}
|
||||
|
||||
// Dynamic
|
||||
|
||||
void argon2d_dyn_hash( void *output, const void *input )
|
||||
{
|
||||
argon2_context context;
|
||||
context.out = (uint8_t *)output;
|
||||
context.outlen = (uint32_t)OUTPUT_BYTES;
|
||||
context.pwd = (uint8_t *)input;
|
||||
context.pwdlen = (uint32_t)INPUT_BYTES;
|
||||
context.salt = (uint8_t *)input; //salt = input
|
||||
context.saltlen = (uint32_t)INPUT_BYTES;
|
||||
context.secret = NULL;
|
||||
context.secretlen = 0;
|
||||
context.ad = NULL;
|
||||
context.adlen = 0;
|
||||
context.allocate_cbk = NULL;
|
||||
context.free_cbk = NULL;
|
||||
context.flags = DEFAULT_ARGON2_FLAG; // = ARGON2_DEFAULT_FLAGS
|
||||
// main configurable Argon2 hash parameters
|
||||
context.m_cost = 500; // Memory in KiB (512KB)
|
||||
context.lanes = 8; // Degree of Parallelism
|
||||
context.threads = 1; // Threads
|
||||
context.t_cost = 2; // Iterations
|
||||
context.version = ARGON2_VERSION_10;
|
||||
|
||||
argon2_ctx( &context, Argon2_d );
|
||||
}
|
||||
|
||||
int scanhash_argon2d_dyn( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done )
|
||||
{
|
||||
uint32_t _ALIGN(64) endiandata[20];
|
||||
uint32_t _ALIGN(64) hash[8];
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
|
||||
uint32_t nonce = first_nonce;
|
||||
|
||||
swab32_array( endiandata, pdata, 20 );
|
||||
|
||||
do {
|
||||
be32enc(&endiandata[19], nonce);
|
||||
argon2d_dyn_hash( hash, endiandata );
|
||||
if ( hash[7] <= Htarg && fulltest( hash, ptarget ) )
|
||||
{
|
||||
pdata[19] = nonce;
|
||||
*hashes_done = pdata[19] - first_nonce;
|
||||
work_set_target_ratio(work, hash);
|
||||
return 1;
|
||||
}
|
||||
nonce++;
|
||||
} while (nonce < max_nonce && !work_restart[thr_id].restart);
|
||||
|
||||
pdata[19] = nonce;
|
||||
*hashes_done = pdata[19] - first_nonce + 1;
|
||||
return 0;
|
||||
}
|
||||
|
||||
bool register_argon2d_dyn_algo( algo_gate_t* gate )
|
||||
{
|
||||
gate->scanhash = (void*)&scanhash_argon2d_dyn;
|
||||
gate->hash = (void*)&argon2d_dyn_hash;
|
||||
gate->set_target = (void*)&scrypt_set_target;
|
||||
gate->optimizations = SSE2_OPT | AVX2_OPT | AVX512_OPT;
|
||||
return true;
|
||||
}
|
||||
|
||||
// Unitus
|
||||
|
||||
int scanhash_argon2d4096( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done)
|
||||
{
|
||||
uint32_t _ALIGN(64) vhash[8];
|
||||
uint32_t _ALIGN(64) endiandata[20];
|
||||
uint32_t *pdata = work->data;
|
||||
uint32_t *ptarget = work->target;
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
const uint32_t first_nonce = pdata[19];
|
||||
uint32_t n = first_nonce;
|
||||
|
||||
uint32_t t_cost = 1; // 1 iteration
|
||||
uint32_t m_cost = 4096; // use 4MB
|
||||
uint32_t parallelism = 1; // 1 thread, 2 lanes
|
||||
|
||||
for ( int i = 0; i < 19; i++ )
|
||||
be32enc( &endiandata[i], pdata[i] );
|
||||
|
||||
do {
|
||||
be32enc( &endiandata[19], n );
|
||||
argon2d_hash_raw( t_cost, m_cost, parallelism, (char*) endiandata, 80,
|
||||
(char*) endiandata, 80, (char*) vhash, 32, ARGON2_VERSION_13 );
|
||||
if ( vhash[7] < Htarg && fulltest( vhash, ptarget ) )
|
||||
{
|
||||
*hashes_done = n - first_nonce + 1;
|
||||
pdata[19] = n;
|
||||
return true;
|
||||
}
|
||||
n++;
|
||||
|
||||
} while (n < max_nonce && !work_restart[thr_id].restart);
|
||||
|
||||
*hashes_done = n - first_nonce + 1;
|
||||
pdata[19] = n;
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
int64_t get_max64_0x1ff() { return 0x1ff; }
|
||||
|
||||
bool register_argon2d4096_algo( algo_gate_t* gate )
|
||||
{
|
||||
gate->scanhash = (void*)&scanhash_argon2d4096;
|
||||
gate->set_target = (void*)&scrypt_set_target;
|
||||
gate->get_max64 = (void*)&get_max64_0x1ff;
|
||||
gate->optimizations = SSE2_OPT | AVX2_OPT | AVX512_OPT;
|
||||
return true;
|
||||
}
|
||||
|
||||
31
algo/argon2/argon2d/argon2d-gate.h
Normal file
31
algo/argon2/argon2d/argon2d-gate.h
Normal file
@@ -0,0 +1,31 @@
|
||||
#ifndef ARGON2D_GATE_H__
|
||||
#define ARGON2D_GATE_H__
|
||||
|
||||
#include "algo-gate-api.h"
|
||||
#include <stdint.h>
|
||||
|
||||
// Credits: version = 0x10, m_cost = 250.
|
||||
bool register_argon2d_crds_algo( algo_gate_t* gate );
|
||||
|
||||
void argon2d_crds_hash( void *state, const void *input );
|
||||
|
||||
int scanhash_argon2d_crds( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done );
|
||||
|
||||
// Dynamic: version = 0x10, m_cost = 500.
|
||||
bool register_argon2d_dyn_algo( algo_gate_t* gate );
|
||||
|
||||
void argon2d_dyn_hash( void *state, const void *input );
|
||||
|
||||
int scanhash_argon2d_dyn( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done );
|
||||
|
||||
|
||||
// Unitus: version = 0x13, m_cost = 4096.
|
||||
bool register_argon2d4096_algo( algo_gate_t* gate );
|
||||
|
||||
int scanhash_argon2d4096( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done );
|
||||
|
||||
#endif
|
||||
|
||||
458
algo/argon2/argon2d/argon2d/argon2.c
Normal file
458
algo/argon2/argon2d/argon2d/argon2.c
Normal file
@@ -0,0 +1,458 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#include <string.h>
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
|
||||
#include "argon2.h"
|
||||
#include "encoding.h"
|
||||
#include "core.h"
|
||||
|
||||
const char *argon2_type2string(argon2_type type, int uppercase) {
|
||||
switch (type) {
|
||||
case Argon2_d:
|
||||
return uppercase ? "Argon2d" : "argon2d";
|
||||
case Argon2_i:
|
||||
return uppercase ? "Argon2i" : "argon2i";
|
||||
case Argon2_id:
|
||||
return uppercase ? "Argon2id" : "argon2id";
|
||||
}
|
||||
|
||||
return NULL;
|
||||
}
|
||||
|
||||
int argon2_ctx(argon2_context *context, argon2_type type) {
|
||||
/* 1. Validate all inputs */
|
||||
int result = validate_inputs(context);
|
||||
uint32_t memory_blocks, segment_length;
|
||||
argon2_instance_t instance;
|
||||
|
||||
if (ARGON2_OK != result) {
|
||||
return result;
|
||||
}
|
||||
|
||||
if (Argon2_d != type && Argon2_i != type && Argon2_id != type) {
|
||||
return ARGON2_INCORRECT_TYPE;
|
||||
}
|
||||
|
||||
/* 2. Align memory size */
|
||||
/* Minimum memory_blocks = 8L blocks, where L is the number of lanes */
|
||||
memory_blocks = context->m_cost;
|
||||
|
||||
if (memory_blocks < 2 * ARGON2_SYNC_POINTS * context->lanes) {
|
||||
memory_blocks = 2 * ARGON2_SYNC_POINTS * context->lanes;
|
||||
}
|
||||
|
||||
segment_length = memory_blocks / (context->lanes * ARGON2_SYNC_POINTS);
|
||||
/* Ensure that all segments have equal length */
|
||||
memory_blocks = segment_length * (context->lanes * ARGON2_SYNC_POINTS);
|
||||
|
||||
instance.version = context->version;
|
||||
instance.memory = NULL;
|
||||
instance.passes = context->t_cost;
|
||||
instance.memory_blocks = memory_blocks;
|
||||
instance.segment_length = segment_length;
|
||||
instance.lane_length = segment_length * ARGON2_SYNC_POINTS;
|
||||
instance.lanes = context->lanes;
|
||||
instance.threads = context->threads;
|
||||
instance.type = type;
|
||||
|
||||
if (instance.threads > instance.lanes) {
|
||||
instance.threads = instance.lanes;
|
||||
}
|
||||
|
||||
/* 3. Initialization: Hashing inputs, allocating memory, filling first
|
||||
* blocks
|
||||
*/
|
||||
result = initialize(&instance, context);
|
||||
|
||||
if (ARGON2_OK != result) {
|
||||
return result;
|
||||
}
|
||||
|
||||
/* 4. Filling memory */
|
||||
result = fill_memory_blocks(&instance);
|
||||
|
||||
if (ARGON2_OK != result) {
|
||||
return result;
|
||||
}
|
||||
/* 5. Finalization */
|
||||
finalize(context, &instance);
|
||||
|
||||
return ARGON2_OK;
|
||||
}
|
||||
|
||||
int argon2_hash(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt, const size_t saltlen,
|
||||
void *hash, const size_t hashlen, char *encoded,
|
||||
const size_t encodedlen, argon2_type type,
|
||||
const uint32_t version){
|
||||
|
||||
argon2_context context;
|
||||
int result;
|
||||
uint8_t *out;
|
||||
|
||||
if (pwdlen > ARGON2_MAX_PWD_LENGTH) {
|
||||
return ARGON2_PWD_TOO_LONG;
|
||||
}
|
||||
|
||||
if (saltlen > ARGON2_MAX_SALT_LENGTH) {
|
||||
return ARGON2_SALT_TOO_LONG;
|
||||
}
|
||||
|
||||
if (hashlen > ARGON2_MAX_OUTLEN) {
|
||||
return ARGON2_OUTPUT_TOO_LONG;
|
||||
}
|
||||
|
||||
if (hashlen < ARGON2_MIN_OUTLEN) {
|
||||
return ARGON2_OUTPUT_TOO_SHORT;
|
||||
}
|
||||
|
||||
out = malloc(hashlen);
|
||||
if (!out) {
|
||||
return ARGON2_MEMORY_ALLOCATION_ERROR;
|
||||
}
|
||||
|
||||
context.out = (uint8_t *)out;
|
||||
context.outlen = (uint32_t)hashlen;
|
||||
context.pwd = CONST_CAST(uint8_t *)pwd;
|
||||
context.pwdlen = (uint32_t)pwdlen;
|
||||
context.salt = CONST_CAST(uint8_t *)salt;
|
||||
context.saltlen = (uint32_t)saltlen;
|
||||
context.secret = NULL;
|
||||
context.secretlen = 0;
|
||||
context.ad = NULL;
|
||||
context.adlen = 0;
|
||||
context.t_cost = t_cost;
|
||||
context.m_cost = m_cost;
|
||||
context.lanes = parallelism;
|
||||
context.threads = parallelism;
|
||||
context.allocate_cbk = NULL;
|
||||
context.free_cbk = NULL;
|
||||
context.flags = ARGON2_DEFAULT_FLAGS;
|
||||
context.version = version;
|
||||
|
||||
result = argon2_ctx(&context, type);
|
||||
|
||||
if (result != ARGON2_OK) {
|
||||
clear_internal_memory(out, hashlen);
|
||||
free(out);
|
||||
return result;
|
||||
}
|
||||
|
||||
/* if raw hash requested, write it */
|
||||
if (hash) {
|
||||
memcpy(hash, out, hashlen);
|
||||
}
|
||||
|
||||
/* if encoding requested, write it */
|
||||
if (encoded && encodedlen) {
|
||||
if (encode_string(encoded, encodedlen, &context, type) != ARGON2_OK) {
|
||||
clear_internal_memory(out, hashlen); /* wipe buffers if error */
|
||||
clear_internal_memory(encoded, encodedlen);
|
||||
free(out);
|
||||
return ARGON2_ENCODING_FAIL;
|
||||
}
|
||||
}
|
||||
clear_internal_memory(out, hashlen);
|
||||
free(out);
|
||||
|
||||
return ARGON2_OK;
|
||||
}
|
||||
|
||||
int argon2i_hash_encoded(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, const size_t hashlen,
|
||||
char *encoded, const size_t encodedlen,
|
||||
const uint32_t version) {
|
||||
|
||||
return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
|
||||
NULL, hashlen, encoded, encodedlen, Argon2_i,
|
||||
version );
|
||||
}
|
||||
|
||||
int argon2i_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, void *hash, const size_t hashlen,
|
||||
const uint32_t version ) {
|
||||
|
||||
return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
|
||||
hash, hashlen, NULL, 0, Argon2_i, version );
|
||||
}
|
||||
|
||||
int argon2d_hash_encoded(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, const size_t hashlen,
|
||||
char *encoded, const size_t encodedlen,
|
||||
const uint32_t version ) {
|
||||
|
||||
return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
|
||||
NULL, hashlen, encoded, encodedlen, Argon2_d,
|
||||
version );
|
||||
}
|
||||
|
||||
int argon2d_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, void *hash, const size_t hashlen,
|
||||
const uint32_t version ) {
|
||||
|
||||
return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
|
||||
hash, hashlen, NULL, 0, Argon2_d, version );
|
||||
}
|
||||
|
||||
int argon2id_hash_encoded(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, const size_t hashlen,
|
||||
char *encoded, const size_t encodedlen,
|
||||
const uint32_t version ) {
|
||||
|
||||
return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
|
||||
NULL, hashlen, encoded, encodedlen, Argon2_id,
|
||||
version);
|
||||
}
|
||||
|
||||
int argon2id_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, void *hash, const size_t hashlen,
|
||||
const uint32_t version ) {
|
||||
return argon2_hash(t_cost, m_cost, parallelism, pwd, pwdlen, salt, saltlen,
|
||||
hash, hashlen, NULL, 0, Argon2_id, version );
|
||||
}
|
||||
|
||||
static int argon2_compare(const uint8_t *b1, const uint8_t *b2, size_t len) {
|
||||
size_t i;
|
||||
uint8_t d = 0U;
|
||||
|
||||
for (i = 0U; i < len; i++) {
|
||||
d |= b1[i] ^ b2[i];
|
||||
}
|
||||
return (int)((1 & ((d - 1) >> 8)) - 1);
|
||||
}
|
||||
|
||||
int argon2_verify(const char *encoded, const void *pwd, const size_t pwdlen,
|
||||
argon2_type type) {
|
||||
|
||||
argon2_context ctx;
|
||||
uint8_t *desired_result = NULL;
|
||||
|
||||
int ret = ARGON2_OK;
|
||||
|
||||
size_t encoded_len;
|
||||
uint32_t max_field_len;
|
||||
|
||||
if (pwdlen > ARGON2_MAX_PWD_LENGTH) {
|
||||
return ARGON2_PWD_TOO_LONG;
|
||||
}
|
||||
|
||||
if (encoded == NULL) {
|
||||
return ARGON2_DECODING_FAIL;
|
||||
}
|
||||
|
||||
encoded_len = strlen(encoded);
|
||||
if (encoded_len > UINT32_MAX) {
|
||||
return ARGON2_DECODING_FAIL;
|
||||
}
|
||||
|
||||
/* No field can be longer than the encoded length */
|
||||
max_field_len = (uint32_t)encoded_len;
|
||||
|
||||
ctx.saltlen = max_field_len;
|
||||
ctx.outlen = max_field_len;
|
||||
|
||||
ctx.salt = malloc(ctx.saltlen);
|
||||
ctx.out = malloc(ctx.outlen);
|
||||
if (!ctx.salt || !ctx.out) {
|
||||
ret = ARGON2_MEMORY_ALLOCATION_ERROR;
|
||||
goto fail;
|
||||
}
|
||||
|
||||
ctx.pwd = (uint8_t *)pwd;
|
||||
ctx.pwdlen = (uint32_t)pwdlen;
|
||||
|
||||
ret = decode_string(&ctx, encoded, type);
|
||||
if (ret != ARGON2_OK) {
|
||||
goto fail;
|
||||
}
|
||||
|
||||
/* Set aside the desired result, and get a new buffer. */
|
||||
desired_result = ctx.out;
|
||||
ctx.out = malloc(ctx.outlen);
|
||||
if (!ctx.out) {
|
||||
ret = ARGON2_MEMORY_ALLOCATION_ERROR;
|
||||
goto fail;
|
||||
}
|
||||
|
||||
ret = argon2_verify_ctx(&ctx, (char *)desired_result, type);
|
||||
if (ret != ARGON2_OK) {
|
||||
goto fail;
|
||||
}
|
||||
|
||||
fail:
|
||||
free(ctx.salt);
|
||||
free(ctx.out);
|
||||
free(desired_result);
|
||||
|
||||
return ret;
|
||||
}
|
||||
|
||||
int argon2i_verify(const char *encoded, const void *pwd, const size_t pwdlen) {
|
||||
|
||||
return argon2_verify(encoded, pwd, pwdlen, Argon2_i);
|
||||
}
|
||||
|
||||
int argon2d_verify(const char *encoded, const void *pwd, const size_t pwdlen) {
|
||||
|
||||
return argon2_verify(encoded, pwd, pwdlen, Argon2_d);
|
||||
}
|
||||
|
||||
int argon2id_verify(const char *encoded, const void *pwd, const size_t pwdlen) {
|
||||
|
||||
return argon2_verify(encoded, pwd, pwdlen, Argon2_id);
|
||||
}
|
||||
|
||||
int argon2d_ctx(argon2_context *context) {
|
||||
return argon2_ctx(context, Argon2_d);
|
||||
}
|
||||
|
||||
int argon2i_ctx(argon2_context *context) {
|
||||
return argon2_ctx(context, Argon2_i);
|
||||
}
|
||||
|
||||
int argon2id_ctx(argon2_context *context) {
|
||||
return argon2_ctx(context, Argon2_id);
|
||||
}
|
||||
|
||||
int argon2_verify_ctx(argon2_context *context, const char *hash,
|
||||
argon2_type type) {
|
||||
int ret = argon2_ctx(context, type);
|
||||
if (ret != ARGON2_OK) {
|
||||
return ret;
|
||||
}
|
||||
|
||||
if (argon2_compare((uint8_t *)hash, context->out, context->outlen)) {
|
||||
return ARGON2_VERIFY_MISMATCH;
|
||||
}
|
||||
|
||||
return ARGON2_OK;
|
||||
}
|
||||
|
||||
int argon2d_verify_ctx(argon2_context *context, const char *hash) {
|
||||
return argon2_verify_ctx(context, hash, Argon2_d);
|
||||
}
|
||||
|
||||
int argon2i_verify_ctx(argon2_context *context, const char *hash) {
|
||||
return argon2_verify_ctx(context, hash, Argon2_i);
|
||||
}
|
||||
|
||||
int argon2id_verify_ctx(argon2_context *context, const char *hash) {
|
||||
return argon2_verify_ctx(context, hash, Argon2_id);
|
||||
}
|
||||
|
||||
const char *argon2_error_message(int error_code) {
|
||||
switch (error_code) {
|
||||
case ARGON2_OK:
|
||||
return "OK";
|
||||
case ARGON2_OUTPUT_PTR_NULL:
|
||||
return "Output pointer is NULL";
|
||||
case ARGON2_OUTPUT_TOO_SHORT:
|
||||
return "Output is too short";
|
||||
case ARGON2_OUTPUT_TOO_LONG:
|
||||
return "Output is too long";
|
||||
case ARGON2_PWD_TOO_SHORT:
|
||||
return "Password is too short";
|
||||
case ARGON2_PWD_TOO_LONG:
|
||||
return "Password is too long";
|
||||
case ARGON2_SALT_TOO_SHORT:
|
||||
return "Salt is too short";
|
||||
case ARGON2_SALT_TOO_LONG:
|
||||
return "Salt is too long";
|
||||
case ARGON2_AD_TOO_SHORT:
|
||||
return "Associated data is too short";
|
||||
case ARGON2_AD_TOO_LONG:
|
||||
return "Associated data is too long";
|
||||
case ARGON2_SECRET_TOO_SHORT:
|
||||
return "Secret is too short";
|
||||
case ARGON2_SECRET_TOO_LONG:
|
||||
return "Secret is too long";
|
||||
case ARGON2_TIME_TOO_SMALL:
|
||||
return "Time cost is too small";
|
||||
case ARGON2_TIME_TOO_LARGE:
|
||||
return "Time cost is too large";
|
||||
case ARGON2_MEMORY_TOO_LITTLE:
|
||||
return "Memory cost is too small";
|
||||
case ARGON2_MEMORY_TOO_MUCH:
|
||||
return "Memory cost is too large";
|
||||
case ARGON2_LANES_TOO_FEW:
|
||||
return "Too few lanes";
|
||||
case ARGON2_LANES_TOO_MANY:
|
||||
return "Too many lanes";
|
||||
case ARGON2_PWD_PTR_MISMATCH:
|
||||
return "Password pointer is NULL, but password length is not 0";
|
||||
case ARGON2_SALT_PTR_MISMATCH:
|
||||
return "Salt pointer is NULL, but salt length is not 0";
|
||||
case ARGON2_SECRET_PTR_MISMATCH:
|
||||
return "Secret pointer is NULL, but secret length is not 0";
|
||||
case ARGON2_AD_PTR_MISMATCH:
|
||||
return "Associated data pointer is NULL, but ad length is not 0";
|
||||
case ARGON2_MEMORY_ALLOCATION_ERROR:
|
||||
return "Memory allocation error";
|
||||
case ARGON2_FREE_MEMORY_CBK_NULL:
|
||||
return "The free memory callback is NULL";
|
||||
case ARGON2_ALLOCATE_MEMORY_CBK_NULL:
|
||||
return "The allocate memory callback is NULL";
|
||||
case ARGON2_INCORRECT_PARAMETER:
|
||||
return "Argon2_Context context is NULL";
|
||||
case ARGON2_INCORRECT_TYPE:
|
||||
return "There is no such version of Argon2";
|
||||
case ARGON2_OUT_PTR_MISMATCH:
|
||||
return "Output pointer mismatch";
|
||||
case ARGON2_THREADS_TOO_FEW:
|
||||
return "Not enough threads";
|
||||
case ARGON2_THREADS_TOO_MANY:
|
||||
return "Too many threads";
|
||||
case ARGON2_MISSING_ARGS:
|
||||
return "Missing arguments";
|
||||
case ARGON2_ENCODING_FAIL:
|
||||
return "Encoding failed";
|
||||
case ARGON2_DECODING_FAIL:
|
||||
return "Decoding failed";
|
||||
case ARGON2_THREAD_FAIL:
|
||||
return "Threading failure";
|
||||
case ARGON2_DECODING_LENGTH_FAIL:
|
||||
return "Some of encoded parameters are too long or too short";
|
||||
case ARGON2_VERIFY_MISMATCH:
|
||||
return "The password does not match the supplied hash";
|
||||
default:
|
||||
return "Unknown error code";
|
||||
}
|
||||
}
|
||||
/*
|
||||
size_t argon2_encodedlen(uint32_t t_cost, uint32_t m_cost, uint32_t parallelism,
|
||||
uint32_t saltlen, uint32_t hashlen, argon2_type type) {
|
||||
return strlen("$$v=$m=,t=,p=$$") + strlen(argon2_type2string(type, 0)) +
|
||||
numlen(t_cost) + numlen(m_cost) + numlen(parallelism) +
|
||||
b64len(saltlen) + b64len(hashlen) + numlen(ARGON2_VERSION_NUMBER) + 1;
|
||||
}
|
||||
*/
|
||||
440
algo/argon2/argon2d/argon2d/argon2.h
Normal file
440
algo/argon2/argon2d/argon2d/argon2.h
Normal file
@@ -0,0 +1,440 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#ifndef ARGON2_H
|
||||
#define ARGON2_H
|
||||
|
||||
#include <stdint.h>
|
||||
#include <stddef.h>
|
||||
#include <limits.h>
|
||||
|
||||
#if defined(__cplusplus)
|
||||
extern "C" {
|
||||
#endif
|
||||
|
||||
/* Symbols visibility control */
|
||||
#ifdef A2_VISCTL
|
||||
#define ARGON2_PUBLIC __attribute__((visibility("default")))
|
||||
#define ARGON2_LOCAL __attribute__ ((visibility ("hidden")))
|
||||
#elif _MSC_VER
|
||||
#define ARGON2_PUBLIC __declspec(dllexport)
|
||||
#define ARGON2_LOCAL
|
||||
#else
|
||||
#define ARGON2_PUBLIC
|
||||
#define ARGON2_LOCAL
|
||||
#endif
|
||||
|
||||
/*
|
||||
* Argon2 input parameter restrictions
|
||||
*/
|
||||
|
||||
/* Minimum and maximum number of lanes (degree of parallelism) */
|
||||
#define ARGON2_MIN_LANES UINT32_C(1)
|
||||
#define ARGON2_MAX_LANES UINT32_C(0xFFFFFF)
|
||||
|
||||
/* Minimum and maximum number of threads */
|
||||
#define ARGON2_MIN_THREADS UINT32_C(1)
|
||||
#define ARGON2_MAX_THREADS UINT32_C(0xFFFFFF)
|
||||
|
||||
/* Number of synchronization points between lanes per pass */
|
||||
#define ARGON2_SYNC_POINTS UINT32_C(4)
|
||||
|
||||
/* Minimum and maximum digest size in bytes */
|
||||
#define ARGON2_MIN_OUTLEN UINT32_C(4)
|
||||
#define ARGON2_MAX_OUTLEN UINT32_C(0xFFFFFFFF)
|
||||
|
||||
/* Minimum and maximum number of memory blocks (each of BLOCK_SIZE bytes) */
|
||||
#define ARGON2_MIN_MEMORY (2 * ARGON2_SYNC_POINTS) /* 2 blocks per slice */
|
||||
|
||||
#define ARGON2_MIN(a, b) ((a) < (b) ? (a) : (b))
|
||||
/* Max memory size is addressing-space/2, topping at 2^32 blocks (4 TB) */
|
||||
#define ARGON2_MAX_MEMORY_BITS \
|
||||
ARGON2_MIN(UINT32_C(32), (sizeof(void *) * CHAR_BIT - 10 - 1))
|
||||
#define ARGON2_MAX_MEMORY \
|
||||
ARGON2_MIN(UINT32_C(0xFFFFFFFF), UINT64_C(1) << ARGON2_MAX_MEMORY_BITS)
|
||||
|
||||
/* Minimum and maximum number of passes */
|
||||
#define ARGON2_MIN_TIME UINT32_C(1)
|
||||
#define ARGON2_MAX_TIME UINT32_C(0xFFFFFFFF)
|
||||
|
||||
/* Minimum and maximum password length in bytes */
|
||||
#define ARGON2_MIN_PWD_LENGTH UINT32_C(0)
|
||||
#define ARGON2_MAX_PWD_LENGTH UINT32_C(0xFFFFFFFF)
|
||||
|
||||
/* Minimum and maximum associated data length in bytes */
|
||||
#define ARGON2_MIN_AD_LENGTH UINT32_C(0)
|
||||
#define ARGON2_MAX_AD_LENGTH UINT32_C(0xFFFFFFFF)
|
||||
|
||||
/* Minimum and maximum salt length in bytes */
|
||||
#define ARGON2_MIN_SALT_LENGTH UINT32_C(8)
|
||||
#define ARGON2_MAX_SALT_LENGTH UINT32_C(0xFFFFFFFF)
|
||||
|
||||
/* Minimum and maximum key length in bytes */
|
||||
#define ARGON2_MIN_SECRET UINT32_C(0)
|
||||
#define ARGON2_MAX_SECRET UINT32_C(0xFFFFFFFF)
|
||||
|
||||
/* Flags to determine which fields are securely wiped (default = no wipe). */
|
||||
#define ARGON2_DEFAULT_FLAGS UINT32_C(0)
|
||||
#define ARGON2_FLAG_CLEAR_PASSWORD (UINT32_C(1) << 0)
|
||||
#define ARGON2_FLAG_CLEAR_SECRET (UINT32_C(1) << 1)
|
||||
|
||||
/* Global flag to determine if we are wiping internal memory buffers. This flag
|
||||
* is defined in core.c and deafults to 1 (wipe internal memory). */
|
||||
extern int FLAG_clear_internal_memory;
|
||||
|
||||
/* Error codes */
|
||||
typedef enum Argon2_ErrorCodes {
|
||||
ARGON2_OK = 0,
|
||||
|
||||
ARGON2_OUTPUT_PTR_NULL = -1,
|
||||
|
||||
ARGON2_OUTPUT_TOO_SHORT = -2,
|
||||
ARGON2_OUTPUT_TOO_LONG = -3,
|
||||
|
||||
ARGON2_PWD_TOO_SHORT = -4,
|
||||
ARGON2_PWD_TOO_LONG = -5,
|
||||
|
||||
ARGON2_SALT_TOO_SHORT = -6,
|
||||
ARGON2_SALT_TOO_LONG = -7,
|
||||
|
||||
ARGON2_AD_TOO_SHORT = -8,
|
||||
ARGON2_AD_TOO_LONG = -9,
|
||||
|
||||
ARGON2_SECRET_TOO_SHORT = -10,
|
||||
ARGON2_SECRET_TOO_LONG = -11,
|
||||
|
||||
ARGON2_TIME_TOO_SMALL = -12,
|
||||
ARGON2_TIME_TOO_LARGE = -13,
|
||||
|
||||
ARGON2_MEMORY_TOO_LITTLE = -14,
|
||||
ARGON2_MEMORY_TOO_MUCH = -15,
|
||||
|
||||
ARGON2_LANES_TOO_FEW = -16,
|
||||
ARGON2_LANES_TOO_MANY = -17,
|
||||
|
||||
ARGON2_PWD_PTR_MISMATCH = -18, /* NULL ptr with non-zero length */
|
||||
ARGON2_SALT_PTR_MISMATCH = -19, /* NULL ptr with non-zero length */
|
||||
ARGON2_SECRET_PTR_MISMATCH = -20, /* NULL ptr with non-zero length */
|
||||
ARGON2_AD_PTR_MISMATCH = -21, /* NULL ptr with non-zero length */
|
||||
|
||||
ARGON2_MEMORY_ALLOCATION_ERROR = -22,
|
||||
|
||||
ARGON2_FREE_MEMORY_CBK_NULL = -23,
|
||||
ARGON2_ALLOCATE_MEMORY_CBK_NULL = -24,
|
||||
|
||||
ARGON2_INCORRECT_PARAMETER = -25,
|
||||
ARGON2_INCORRECT_TYPE = -26,
|
||||
|
||||
ARGON2_OUT_PTR_MISMATCH = -27,
|
||||
|
||||
ARGON2_THREADS_TOO_FEW = -28,
|
||||
ARGON2_THREADS_TOO_MANY = -29,
|
||||
|
||||
ARGON2_MISSING_ARGS = -30,
|
||||
|
||||
ARGON2_ENCODING_FAIL = -31,
|
||||
|
||||
ARGON2_DECODING_FAIL = -32,
|
||||
|
||||
ARGON2_THREAD_FAIL = -33,
|
||||
|
||||
ARGON2_DECODING_LENGTH_FAIL = -34,
|
||||
|
||||
ARGON2_VERIFY_MISMATCH = -35
|
||||
} argon2_error_codes;
|
||||
|
||||
/* Memory allocator types --- for external allocation */
|
||||
typedef int (*allocate_fptr)(uint8_t **memory, size_t bytes_to_allocate);
|
||||
typedef void (*deallocate_fptr)(uint8_t *memory, size_t bytes_to_allocate);
|
||||
|
||||
/* Argon2 external data structures */
|
||||
|
||||
/*
|
||||
*****
|
||||
* Context: structure to hold Argon2 inputs:
|
||||
* output array and its length,
|
||||
* password and its length,
|
||||
* salt and its length,
|
||||
* secret and its length,
|
||||
* associated data and its length,
|
||||
* number of passes, amount of used memory (in KBytes, can be rounded up a bit)
|
||||
* number of parallel threads that will be run.
|
||||
* All the parameters above affect the output hash value.
|
||||
* Additionally, two function pointers can be provided to allocate and
|
||||
* deallocate the memory (if NULL, memory will be allocated internally).
|
||||
* Also, three flags indicate whether to erase password, secret as soon as they
|
||||
* are pre-hashed (and thus not needed anymore), and the entire memory
|
||||
*****
|
||||
* Simplest situation: you have output array out[8], password is stored in
|
||||
* pwd[32], salt is stored in salt[16], you do not have keys nor associated
|
||||
* data. You need to spend 1 GB of RAM and you run 5 passes of Argon2d with
|
||||
* 4 parallel lanes.
|
||||
* You want to erase the password, but you're OK with last pass not being
|
||||
* erased. You want to use the default memory allocator.
|
||||
* Then you initialize:
|
||||
Argon2_Context(out,8,pwd,32,salt,16,NULL,0,NULL,0,5,1<<20,4,4,NULL,NULL,true,false,false,false)
|
||||
*/
|
||||
typedef struct Argon2_Context {
|
||||
uint8_t *out; /* output array */
|
||||
uint32_t outlen; /* digest length */
|
||||
|
||||
uint8_t *pwd; /* password array */
|
||||
uint32_t pwdlen; /* password length */
|
||||
|
||||
uint8_t *salt; /* salt array */
|
||||
uint32_t saltlen; /* salt length */
|
||||
|
||||
uint8_t *secret; /* key array */
|
||||
uint32_t secretlen; /* key length */
|
||||
|
||||
uint8_t *ad; /* associated data array */
|
||||
uint32_t adlen; /* associated data length */
|
||||
|
||||
uint32_t t_cost; /* number of passes */
|
||||
uint32_t m_cost; /* amount of memory requested (KB) */
|
||||
uint32_t lanes; /* number of lanes */
|
||||
uint32_t threads; /* maximum number of threads */
|
||||
|
||||
uint32_t version; /* version number */
|
||||
|
||||
allocate_fptr allocate_cbk; /* pointer to memory allocator */
|
||||
deallocate_fptr free_cbk; /* pointer to memory deallocator */
|
||||
|
||||
uint32_t flags; /* array of bool options */
|
||||
} argon2_context;
|
||||
|
||||
/* Argon2 primitive type */
|
||||
typedef enum Argon2_type {
|
||||
Argon2_d = 0,
|
||||
Argon2_i = 1,
|
||||
Argon2_id = 2
|
||||
} argon2_type;
|
||||
|
||||
/* Version of the algorithm */
|
||||
#define ARGON2_VERSION_10 0x10
|
||||
#define ARGON2_VERSION_13 0x13
|
||||
|
||||
/*
|
||||
* Function that gives the string representation of an argon2_type.
|
||||
* @param type The argon2_type that we want the string for
|
||||
* @param uppercase Whether the string should have the first letter uppercase
|
||||
* @return NULL if invalid type, otherwise the string representation.
|
||||
*/
|
||||
ARGON2_PUBLIC const char *argon2_type2string(argon2_type type, int uppercase);
|
||||
|
||||
/*
|
||||
* Function that performs memory-hard hashing with certain degree of parallelism
|
||||
* @param context Pointer to the Argon2 internal structure
|
||||
* @return Error code if smth is wrong, ARGON2_OK otherwise
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2_ctx(argon2_context *context, argon2_type type);
|
||||
|
||||
/**
|
||||
* Hashes a password with Argon2i, producing an encoded hash
|
||||
* @param t_cost Number of iterations
|
||||
* @param m_cost Sets memory usage to m_cost kibibytes
|
||||
* @param parallelism Number of threads and compute lanes
|
||||
* @param pwd Pointer to password
|
||||
* @param pwdlen Password size in bytes
|
||||
* @param salt Pointer to salt
|
||||
* @param saltlen Salt size in bytes
|
||||
* @param hashlen Desired length of the hash in bytes
|
||||
* @param encoded Buffer where to write the encoded hash
|
||||
* @param encodedlen Size of the buffer (thus max size of the encoded hash)
|
||||
* @pre Different parallelism levels will give different results
|
||||
* @pre Returns ARGON2_OK if successful
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2i_hash_encoded(const uint32_t t_cost,
|
||||
const uint32_t m_cost,
|
||||
const uint32_t parallelism,
|
||||
const void *pwd, const size_t pwdlen,
|
||||
const void *salt, const size_t saltlen,
|
||||
const size_t hashlen, char *encoded,
|
||||
const size_t encodedlen,
|
||||
const uint32_t version );
|
||||
|
||||
/**
|
||||
* Hashes a password with Argon2i, producing a raw hash at @hash
|
||||
* @param t_cost Number of iterations
|
||||
* @param m_cost Sets memory usage to m_cost kibibytes
|
||||
* @param parallelism Number of threads and compute lanes
|
||||
* @param pwd Pointer to password
|
||||
* @param pwdlen Password size in bytes
|
||||
* @param salt Pointer to salt
|
||||
* @param saltlen Salt size in bytes
|
||||
* @param hash Buffer where to write the raw hash - updated by the function
|
||||
* @param hashlen Desired length of the hash in bytes
|
||||
* @pre Different parallelism levels will give different results
|
||||
* @pre Returns ARGON2_OK if successful
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2i_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, void *hash,
|
||||
const size_t hashlen,
|
||||
const uint32_t version );
|
||||
|
||||
ARGON2_PUBLIC int argon2d_hash_encoded(const uint32_t t_cost,
|
||||
const uint32_t m_cost,
|
||||
const uint32_t parallelism,
|
||||
const void *pwd, const size_t pwdlen,
|
||||
const void *salt, const size_t saltlen,
|
||||
const size_t hashlen, char *encoded,
|
||||
const size_t encodedlen,
|
||||
const uint32_t version );
|
||||
|
||||
ARGON2_PUBLIC int argon2d_hash_raw(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, void *hash,
|
||||
const size_t hashlen,
|
||||
const uint32_t version );
|
||||
|
||||
ARGON2_PUBLIC int argon2id_hash_encoded(const uint32_t t_cost,
|
||||
const uint32_t m_cost,
|
||||
const uint32_t parallelism,
|
||||
const void *pwd, const size_t pwdlen,
|
||||
const void *salt, const size_t saltlen,
|
||||
const size_t hashlen, char *encoded,
|
||||
const size_t encodedlen,
|
||||
const uint32_t version );
|
||||
|
||||
ARGON2_PUBLIC int argon2id_hash_raw(const uint32_t t_cost,
|
||||
const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, void *hash,
|
||||
const size_t hashlen,
|
||||
const uint32_t version );
|
||||
|
||||
/* generic function underlying the above ones */
|
||||
ARGON2_PUBLIC int argon2_hash(const uint32_t t_cost, const uint32_t m_cost,
|
||||
const uint32_t parallelism, const void *pwd,
|
||||
const size_t pwdlen, const void *salt,
|
||||
const size_t saltlen, void *hash,
|
||||
const size_t hashlen, char *encoded,
|
||||
const size_t encodedlen, argon2_type type,
|
||||
const uint32_t version );
|
||||
|
||||
/**
|
||||
* Verifies a password against an encoded string
|
||||
* Encoded string is restricted as in validate_inputs()
|
||||
* @param encoded String encoding parameters, salt, hash
|
||||
* @param pwd Pointer to password
|
||||
* @pre Returns ARGON2_OK if successful
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2i_verify(const char *encoded, const void *pwd,
|
||||
const size_t pwdlen);
|
||||
|
||||
ARGON2_PUBLIC int argon2d_verify(const char *encoded, const void *pwd,
|
||||
const size_t pwdlen);
|
||||
|
||||
ARGON2_PUBLIC int argon2id_verify(const char *encoded, const void *pwd,
|
||||
const size_t pwdlen);
|
||||
|
||||
/* generic function underlying the above ones */
|
||||
ARGON2_PUBLIC int argon2_verify(const char *encoded, const void *pwd,
|
||||
const size_t pwdlen, argon2_type type);
|
||||
|
||||
/**
|
||||
* Argon2d: Version of Argon2 that picks memory blocks depending
|
||||
* on the password and salt. Only for side-channel-free
|
||||
* environment!!
|
||||
*****
|
||||
* @param context Pointer to current Argon2 context
|
||||
* @return Zero if successful, a non zero error code otherwise
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2d_ctx(argon2_context *context);
|
||||
|
||||
/**
|
||||
* Argon2i: Version of Argon2 that picks memory blocks
|
||||
* independent on the password and salt. Good for side-channels,
|
||||
* but worse w.r.t. tradeoff attacks if only one pass is used.
|
||||
*****
|
||||
* @param context Pointer to current Argon2 context
|
||||
* @return Zero if successful, a non zero error code otherwise
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2i_ctx(argon2_context *context);
|
||||
|
||||
/**
|
||||
* Argon2id: Version of Argon2 where the first half-pass over memory is
|
||||
* password-independent, the rest are password-dependent (on the password and
|
||||
* salt). OK against side channels (they reduce to 1/2-pass Argon2i), and
|
||||
* better with w.r.t. tradeoff attacks (similar to Argon2d).
|
||||
*****
|
||||
* @param context Pointer to current Argon2 context
|
||||
* @return Zero if successful, a non zero error code otherwise
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2id_ctx(argon2_context *context);
|
||||
|
||||
/**
|
||||
* Verify if a given password is correct for Argon2d hashing
|
||||
* @param context Pointer to current Argon2 context
|
||||
* @param hash The password hash to verify. The length of the hash is
|
||||
* specified by the context outlen member
|
||||
* @return Zero if successful, a non zero error code otherwise
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2d_verify_ctx(argon2_context *context, const char *hash);
|
||||
|
||||
/**
|
||||
* Verify if a given password is correct for Argon2i hashing
|
||||
* @param context Pointer to current Argon2 context
|
||||
* @param hash The password hash to verify. The length of the hash is
|
||||
* specified by the context outlen member
|
||||
* @return Zero if successful, a non zero error code otherwise
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2i_verify_ctx(argon2_context *context, const char *hash);
|
||||
|
||||
/**
|
||||
* Verify if a given password is correct for Argon2id hashing
|
||||
* @param context Pointer to current Argon2 context
|
||||
* @param hash The password hash to verify. The length of the hash is
|
||||
* specified by the context outlen member
|
||||
* @return Zero if successful, a non zero error code otherwise
|
||||
*/
|
||||
ARGON2_PUBLIC int argon2id_verify_ctx(argon2_context *context,
|
||||
const char *hash);
|
||||
|
||||
/* generic function underlying the above ones */
|
||||
ARGON2_PUBLIC int argon2_verify_ctx(argon2_context *context, const char *hash,
|
||||
argon2_type type);
|
||||
|
||||
/**
|
||||
* Get the associated error message for given error code
|
||||
* @return The error message associated with the given error code
|
||||
*/
|
||||
ARGON2_PUBLIC const char *argon2_error_message(int error_code);
|
||||
|
||||
/**
|
||||
* Returns the encoded hash length for the given input parameters
|
||||
* @param t_cost Number of iterations
|
||||
* @param m_cost Memory usage in kibibytes
|
||||
* @param parallelism Number of threads; used to compute lanes
|
||||
* @param saltlen Salt size in bytes
|
||||
* @param hashlen Hash size in bytes
|
||||
* @param type The argon2_type that we want the encoded length for
|
||||
* @return The encoded hash length in bytes
|
||||
*/
|
||||
ARGON2_PUBLIC size_t argon2_encodedlen(uint32_t t_cost, uint32_t m_cost,
|
||||
uint32_t parallelism, uint32_t saltlen,
|
||||
uint32_t hashlen, argon2_type type);
|
||||
|
||||
#if defined(__cplusplus)
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif
|
||||
57
algo/argon2/argon2d/argon2d/argon2d_thread.c
Normal file
57
algo/argon2/argon2d/argon2d/argon2d_thread.c
Normal file
@@ -0,0 +1,57 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#if !defined(ARGON2_NO_THREADS)
|
||||
|
||||
#include "argon2d_thread.h"
|
||||
#if defined(_WIN32)
|
||||
#include <windows.h>
|
||||
#endif
|
||||
|
||||
int argon2_thread_create(argon2_thread_handle_t *handle,
|
||||
argon2_thread_func_t func, void *args) {
|
||||
if (NULL == handle || func == NULL) {
|
||||
return -1;
|
||||
}
|
||||
#if defined(_WIN32)
|
||||
*handle = _beginthreadex(NULL, 0, func, args, 0, NULL);
|
||||
return *handle != 0 ? 0 : -1;
|
||||
#else
|
||||
return pthread_create(handle, NULL, func, args);
|
||||
#endif
|
||||
}
|
||||
|
||||
int argon2_thread_join(argon2_thread_handle_t handle) {
|
||||
#if defined(_WIN32)
|
||||
if (WaitForSingleObject((HANDLE)handle, INFINITE) == WAIT_OBJECT_0) {
|
||||
return CloseHandle((HANDLE)handle) != 0 ? 0 : -1;
|
||||
}
|
||||
return -1;
|
||||
#else
|
||||
return pthread_join(handle, NULL);
|
||||
#endif
|
||||
}
|
||||
|
||||
void argon2_thread_exit(void) {
|
||||
#if defined(_WIN32)
|
||||
_endthreadex(0);
|
||||
#else
|
||||
pthread_exit(NULL);
|
||||
#endif
|
||||
}
|
||||
|
||||
#endif /* ARGON2_NO_THREADS */
|
||||
67
algo/argon2/argon2d/argon2d/argon2d_thread.h
Normal file
67
algo/argon2/argon2d/argon2d/argon2d_thread.h
Normal file
@@ -0,0 +1,67 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#ifndef ARGON2_THREAD_H
|
||||
#define ARGON2_THREAD_H
|
||||
|
||||
#if !defined(ARGON2_NO_THREADS)
|
||||
|
||||
/*
|
||||
Here we implement an abstraction layer for the simpĺe requirements
|
||||
of the Argon2 code. We only require 3 primitives---thread creation,
|
||||
joining, and termination---so full emulation of the pthreads API
|
||||
is unwarranted. Currently we wrap pthreads and Win32 threads.
|
||||
|
||||
The API defines 2 types: the function pointer type,
|
||||
argon2_thread_func_t,
|
||||
and the type of the thread handle---argon2_thread_handle_t.
|
||||
*/
|
||||
#if defined(_WIN32)
|
||||
#include <process.h>
|
||||
typedef unsigned(__stdcall *argon2_thread_func_t)(void *);
|
||||
typedef uintptr_t argon2_thread_handle_t;
|
||||
#else
|
||||
#include <pthread.h>
|
||||
typedef void *(*argon2_thread_func_t)(void *);
|
||||
typedef pthread_t argon2_thread_handle_t;
|
||||
#endif
|
||||
|
||||
/* Creates a thread
|
||||
* @param handle pointer to a thread handle, which is the output of this
|
||||
* function. Must not be NULL.
|
||||
* @param func A function pointer for the thread's entry point. Must not be
|
||||
* NULL.
|
||||
* @param args Pointer that is passed as an argument to @func. May be NULL.
|
||||
* @return 0 if @handle and @func are valid pointers and a thread is successfully
|
||||
* created.
|
||||
*/
|
||||
int argon2_thread_create(argon2_thread_handle_t *handle,
|
||||
argon2_thread_func_t func, void *args);
|
||||
|
||||
/* Waits for a thread to terminate
|
||||
* @param handle Handle to a thread created with argon2_thread_create.
|
||||
* @return 0 if @handle is a valid handle, and joining completed successfully.
|
||||
*/
|
||||
int argon2_thread_join(argon2_thread_handle_t handle);
|
||||
|
||||
/* Terminate the current thread. Must be run inside a thread created by
|
||||
* argon2_thread_create.
|
||||
*/
|
||||
void argon2_thread_exit(void);
|
||||
|
||||
#endif /* ARGON2_NO_THREADS */
|
||||
#endif
|
||||
635
algo/argon2/argon2d/argon2d/core.c
Normal file
635
algo/argon2/argon2d/argon2d/core.c
Normal file
@@ -0,0 +1,635 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
/*For memory wiping*/
|
||||
#ifdef _MSC_VER
|
||||
#include <windows.h>
|
||||
#include <winbase.h> /* For SecureZeroMemory */
|
||||
#endif
|
||||
#if defined __STDC_LIB_EXT1__
|
||||
#define __STDC_WANT_LIB_EXT1__ 1
|
||||
#endif
|
||||
#define VC_GE_2005(version) (version >= 1400)
|
||||
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
#include "core.h"
|
||||
#include "argon2d_thread.h"
|
||||
#include "../blake2/blake2.h"
|
||||
#include "../blake2/blake2-impl.h"
|
||||
|
||||
#ifdef GENKAT
|
||||
#include "genkat.h"
|
||||
#endif
|
||||
|
||||
#if defined(__clang__)
|
||||
#if __has_attribute(optnone)
|
||||
#define NOT_OPTIMIZED __attribute__((optnone))
|
||||
#endif
|
||||
#elif defined(__GNUC__)
|
||||
#define GCC_VERSION \
|
||||
(__GNUC__ * 10000 + __GNUC_MINOR__ * 100 + __GNUC_PATCHLEVEL__)
|
||||
#if GCC_VERSION >= 40400
|
||||
#define NOT_OPTIMIZED __attribute__((optimize("O0")))
|
||||
#endif
|
||||
#endif
|
||||
#ifndef NOT_OPTIMIZED
|
||||
#define NOT_OPTIMIZED
|
||||
#endif
|
||||
|
||||
/***************Instance and Position constructors**********/
|
||||
void init_block_value(block *b, uint8_t in) { memset(b->v, in, sizeof(b->v)); }
|
||||
|
||||
void copy_block(block *dst, const block *src) {
|
||||
memcpy(dst->v, src->v, sizeof(uint64_t) * ARGON2_QWORDS_IN_BLOCK);
|
||||
}
|
||||
|
||||
void xor_block(block *dst, const block *src) {
|
||||
int i;
|
||||
for (i = 0; i < ARGON2_QWORDS_IN_BLOCK; ++i) {
|
||||
dst->v[i] ^= src->v[i];
|
||||
}
|
||||
}
|
||||
|
||||
static void load_block(block *dst, const void *input) {
|
||||
unsigned i;
|
||||
for (i = 0; i < ARGON2_QWORDS_IN_BLOCK; ++i) {
|
||||
dst->v[i] = load64((const uint8_t *)input + i * sizeof(dst->v[i]));
|
||||
}
|
||||
}
|
||||
|
||||
static void store_block(void *output, const block *src) {
|
||||
unsigned i;
|
||||
for (i = 0; i < ARGON2_QWORDS_IN_BLOCK; ++i) {
|
||||
store64((uint8_t *)output + i * sizeof(src->v[i]), src->v[i]);
|
||||
}
|
||||
}
|
||||
|
||||
/***************Memory functions*****************/
|
||||
|
||||
int allocate_memory(const argon2_context *context, uint8_t **memory,
|
||||
size_t num, size_t size) {
|
||||
size_t memory_size = num*size;
|
||||
if (memory == NULL) {
|
||||
return ARGON2_MEMORY_ALLOCATION_ERROR;
|
||||
}
|
||||
|
||||
/* 1. Check for multiplication overflow */
|
||||
if (size != 0 && memory_size / size != num) {
|
||||
return ARGON2_MEMORY_ALLOCATION_ERROR;
|
||||
}
|
||||
|
||||
/* 2. Try to allocate with appropriate allocator */
|
||||
if (context->allocate_cbk) {
|
||||
(context->allocate_cbk)(memory, memory_size);
|
||||
} else {
|
||||
*memory = malloc(memory_size);
|
||||
}
|
||||
|
||||
if (*memory == NULL) {
|
||||
return ARGON2_MEMORY_ALLOCATION_ERROR;
|
||||
}
|
||||
|
||||
return ARGON2_OK;
|
||||
}
|
||||
|
||||
void free_memory(const argon2_context *context, uint8_t *memory,
|
||||
size_t num, size_t size) {
|
||||
size_t memory_size = num*size;
|
||||
// clear_internal_memory(memory, memory_size);
|
||||
if (context->free_cbk) {
|
||||
(context->free_cbk)(memory, memory_size);
|
||||
} else {
|
||||
free(memory);
|
||||
}
|
||||
}
|
||||
|
||||
void NOT_OPTIMIZED secure_wipe_memory(void *v, size_t n) {
|
||||
#if defined(_MSC_VER) && VC_GE_2005(_MSC_VER)
|
||||
SecureZeroMemory(v, n);
|
||||
#elif defined memset_s
|
||||
memset_s(v, n, 0, n);
|
||||
#elif defined(__OpenBSD__)
|
||||
explicit_bzero(v, n);
|
||||
#else
|
||||
static void *(*const volatile memset_sec)(void *, int, size_t) = &memset;
|
||||
memset_sec(v, 0, n);
|
||||
#endif
|
||||
}
|
||||
|
||||
/* Memory clear flag defaults to true. */
|
||||
int FLAG_clear_internal_memory = 0;
|
||||
void clear_internal_memory(void *v, size_t n) {
|
||||
if (FLAG_clear_internal_memory && v) {
|
||||
// secure_wipe_memory(v, n);
|
||||
}
|
||||
}
|
||||
|
||||
void finalize(const argon2_context *context, argon2_instance_t *instance) {
|
||||
if (context != NULL && instance != NULL) {
|
||||
block blockhash;
|
||||
uint32_t l;
|
||||
|
||||
copy_block(&blockhash, instance->memory + instance->lane_length - 1);
|
||||
|
||||
/* XOR the last blocks */
|
||||
for (l = 1; l < instance->lanes; ++l) {
|
||||
uint32_t last_block_in_lane =
|
||||
l * instance->lane_length + (instance->lane_length - 1);
|
||||
xor_block(&blockhash, instance->memory + last_block_in_lane);
|
||||
}
|
||||
|
||||
/* Hash the result */
|
||||
{
|
||||
uint8_t blockhash_bytes[ARGON2_BLOCK_SIZE];
|
||||
store_block(blockhash_bytes, &blockhash);
|
||||
blake2b_long(context->out, context->outlen, blockhash_bytes,
|
||||
ARGON2_BLOCK_SIZE);
|
||||
/* clear blockhash and blockhash_bytes */
|
||||
clear_internal_memory(blockhash.v, ARGON2_BLOCK_SIZE);
|
||||
clear_internal_memory(blockhash_bytes, ARGON2_BLOCK_SIZE);
|
||||
}
|
||||
|
||||
#ifdef GENKAT
|
||||
print_tag(context->out, context->outlen);
|
||||
#endif
|
||||
|
||||
free_memory(context, (uint8_t *)instance->memory,
|
||||
instance->memory_blocks, sizeof(block));
|
||||
}
|
||||
}
|
||||
|
||||
uint32_t index_alpha(const argon2_instance_t *instance,
|
||||
const argon2_position_t *position, uint32_t pseudo_rand,
|
||||
int same_lane) {
|
||||
/*
|
||||
* Pass 0:
|
||||
* This lane : all already finished segments plus already constructed
|
||||
* blocks in this segment
|
||||
* Other lanes : all already finished segments
|
||||
* Pass 1+:
|
||||
* This lane : (SYNC_POINTS - 1) last segments plus already constructed
|
||||
* blocks in this segment
|
||||
* Other lanes : (SYNC_POINTS - 1) last segments
|
||||
*/
|
||||
uint32_t reference_area_size;
|
||||
uint64_t relative_position;
|
||||
uint32_t start_position, absolute_position;
|
||||
|
||||
if (0 == position->pass) {
|
||||
/* First pass */
|
||||
if (0 == position->slice) {
|
||||
/* First slice */
|
||||
reference_area_size =
|
||||
position->index - 1; /* all but the previous */
|
||||
} else {
|
||||
if (same_lane) {
|
||||
/* The same lane => add current segment */
|
||||
reference_area_size =
|
||||
position->slice * instance->segment_length +
|
||||
position->index - 1;
|
||||
} else {
|
||||
reference_area_size =
|
||||
position->slice * instance->segment_length +
|
||||
((position->index == 0) ? (-1) : 0);
|
||||
}
|
||||
}
|
||||
} else {
|
||||
/* Second pass */
|
||||
if (same_lane) {
|
||||
reference_area_size = instance->lane_length -
|
||||
instance->segment_length + position->index -
|
||||
1;
|
||||
} else {
|
||||
reference_area_size = instance->lane_length -
|
||||
instance->segment_length +
|
||||
((position->index == 0) ? (-1) : 0);
|
||||
}
|
||||
}
|
||||
|
||||
/* 1.2.4. Mapping pseudo_rand to 0..<reference_area_size-1> and produce
|
||||
* relative position */
|
||||
relative_position = pseudo_rand;
|
||||
relative_position = relative_position * relative_position >> 32;
|
||||
relative_position = reference_area_size - 1 -
|
||||
(reference_area_size * relative_position >> 32);
|
||||
|
||||
/* 1.2.5 Computing starting position */
|
||||
start_position = 0;
|
||||
|
||||
if (0 != position->pass) {
|
||||
start_position = (position->slice == ARGON2_SYNC_POINTS - 1)
|
||||
? 0
|
||||
: (position->slice + 1) * instance->segment_length;
|
||||
}
|
||||
|
||||
/* 1.2.6. Computing absolute position */
|
||||
absolute_position = (start_position + relative_position) %
|
||||
instance->lane_length; /* absolute position */
|
||||
return absolute_position;
|
||||
}
|
||||
|
||||
/* Single-threaded version for p=1 case */
|
||||
static int fill_memory_blocks_st(argon2_instance_t *instance) {
|
||||
uint32_t r, s, l;
|
||||
|
||||
for (r = 0; r < instance->passes; ++r) {
|
||||
for (s = 0; s < ARGON2_SYNC_POINTS; ++s) {
|
||||
for (l = 0; l < instance->lanes; ++l) {
|
||||
argon2_position_t position = {r, l, (uint8_t)s, 0};
|
||||
fill_segment(instance, position);
|
||||
}
|
||||
}
|
||||
#ifdef GENKAT
|
||||
internal_kat(instance, r); /* Print all memory blocks */
|
||||
#endif
|
||||
}
|
||||
return ARGON2_OK;
|
||||
}
|
||||
|
||||
#if !defined(ARGON2_NO_THREADS)
|
||||
|
||||
#ifdef _WIN32
|
||||
static unsigned __stdcall fill_segment_thr(void *thread_data)
|
||||
#else
|
||||
static void *fill_segment_thr(void *thread_data)
|
||||
#endif
|
||||
{
|
||||
argon2_thread_data *my_data = thread_data;
|
||||
fill_segment(my_data->instance_ptr, my_data->pos);
|
||||
argon2_thread_exit();
|
||||
return 0;
|
||||
}
|
||||
|
||||
/* Multi-threaded version for p > 1 case */
|
||||
static int fill_memory_blocks_mt(argon2_instance_t *instance) {
|
||||
uint32_t r, s;
|
||||
argon2_thread_handle_t *thread = NULL;
|
||||
argon2_thread_data *thr_data = NULL;
|
||||
int rc = ARGON2_OK;
|
||||
|
||||
/* 1. Allocating space for threads */
|
||||
thread = calloc(instance->lanes, sizeof(argon2_thread_handle_t));
|
||||
if (thread == NULL) {
|
||||
rc = ARGON2_MEMORY_ALLOCATION_ERROR;
|
||||
goto fail;
|
||||
}
|
||||
|
||||
thr_data = calloc(instance->lanes, sizeof(argon2_thread_data));
|
||||
if (thr_data == NULL) {
|
||||
rc = ARGON2_MEMORY_ALLOCATION_ERROR;
|
||||
goto fail;
|
||||
}
|
||||
|
||||
for (r = 0; r < instance->passes; ++r) {
|
||||
for (s = 0; s < ARGON2_SYNC_POINTS; ++s) {
|
||||
uint32_t l;
|
||||
|
||||
/* 2. Calling threads */
|
||||
for (l = 0; l < instance->lanes; ++l) {
|
||||
argon2_position_t position;
|
||||
|
||||
/* 2.1 Join a thread if limit is exceeded */
|
||||
if (l >= instance->threads) {
|
||||
if (argon2_thread_join(thread[l - instance->threads])) {
|
||||
rc = ARGON2_THREAD_FAIL;
|
||||
goto fail;
|
||||
}
|
||||
}
|
||||
|
||||
/* 2.2 Create thread */
|
||||
position.pass = r;
|
||||
position.lane = l;
|
||||
position.slice = (uint8_t)s;
|
||||
position.index = 0;
|
||||
thr_data[l].instance_ptr =
|
||||
instance; /* preparing the thread input */
|
||||
memcpy(&(thr_data[l].pos), &position,
|
||||
sizeof(argon2_position_t));
|
||||
if (argon2_thread_create(&thread[l], &fill_segment_thr,
|
||||
(void *)&thr_data[l])) {
|
||||
rc = ARGON2_THREAD_FAIL;
|
||||
goto fail;
|
||||
}
|
||||
|
||||
/* fill_segment(instance, position); */
|
||||
/*Non-thread equivalent of the lines above */
|
||||
}
|
||||
|
||||
/* 3. Joining remaining threads */
|
||||
for (l = instance->lanes - instance->threads; l < instance->lanes;
|
||||
++l) {
|
||||
if (argon2_thread_join(thread[l])) {
|
||||
rc = ARGON2_THREAD_FAIL;
|
||||
goto fail;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
#ifdef GENKAT
|
||||
internal_kat(instance, r); /* Print all memory blocks */
|
||||
#endif
|
||||
}
|
||||
|
||||
fail:
|
||||
if (thread != NULL) {
|
||||
free(thread);
|
||||
}
|
||||
if (thr_data != NULL) {
|
||||
free(thr_data);
|
||||
}
|
||||
return rc;
|
||||
}
|
||||
|
||||
#endif /* ARGON2_NO_THREADS */
|
||||
|
||||
int fill_memory_blocks(argon2_instance_t *instance) {
|
||||
if (instance == NULL || instance->lanes == 0) {
|
||||
return ARGON2_INCORRECT_PARAMETER;
|
||||
}
|
||||
#if defined(ARGON2_NO_THREADS)
|
||||
return fill_memory_blocks_st(instance);
|
||||
#else
|
||||
return instance->threads == 1 ?
|
||||
fill_memory_blocks_st(instance) : fill_memory_blocks_mt(instance);
|
||||
#endif
|
||||
}
|
||||
|
||||
int validate_inputs(const argon2_context *context) {
|
||||
if (NULL == context) {
|
||||
return ARGON2_INCORRECT_PARAMETER;
|
||||
}
|
||||
|
||||
if (NULL == context->out) {
|
||||
return ARGON2_OUTPUT_PTR_NULL;
|
||||
}
|
||||
|
||||
/* Validate output length */
|
||||
if (ARGON2_MIN_OUTLEN > context->outlen) {
|
||||
return ARGON2_OUTPUT_TOO_SHORT;
|
||||
}
|
||||
|
||||
if (ARGON2_MAX_OUTLEN < context->outlen) {
|
||||
return ARGON2_OUTPUT_TOO_LONG;
|
||||
}
|
||||
|
||||
/* Validate password (required param) */
|
||||
if (NULL == context->pwd) {
|
||||
if (0 != context->pwdlen) {
|
||||
return ARGON2_PWD_PTR_MISMATCH;
|
||||
}
|
||||
}
|
||||
|
||||
if (ARGON2_MIN_PWD_LENGTH > context->pwdlen) {
|
||||
return ARGON2_PWD_TOO_SHORT;
|
||||
}
|
||||
|
||||
if (ARGON2_MAX_PWD_LENGTH < context->pwdlen) {
|
||||
return ARGON2_PWD_TOO_LONG;
|
||||
}
|
||||
|
||||
/* Validate salt (required param) */
|
||||
if (NULL == context->salt) {
|
||||
if (0 != context->saltlen) {
|
||||
return ARGON2_SALT_PTR_MISMATCH;
|
||||
}
|
||||
}
|
||||
|
||||
if (ARGON2_MIN_SALT_LENGTH > context->saltlen) {
|
||||
return ARGON2_SALT_TOO_SHORT;
|
||||
}
|
||||
|
||||
if (ARGON2_MAX_SALT_LENGTH < context->saltlen) {
|
||||
return ARGON2_SALT_TOO_LONG;
|
||||
}
|
||||
|
||||
/* Validate secret (optional param) */
|
||||
if (NULL == context->secret) {
|
||||
if (0 != context->secretlen) {
|
||||
return ARGON2_SECRET_PTR_MISMATCH;
|
||||
}
|
||||
} else {
|
||||
if (ARGON2_MIN_SECRET > context->secretlen) {
|
||||
return ARGON2_SECRET_TOO_SHORT;
|
||||
}
|
||||
if (ARGON2_MAX_SECRET < context->secretlen) {
|
||||
return ARGON2_SECRET_TOO_LONG;
|
||||
}
|
||||
}
|
||||
|
||||
/* Validate associated data (optional param) */
|
||||
if (NULL == context->ad) {
|
||||
if (0 != context->adlen) {
|
||||
return ARGON2_AD_PTR_MISMATCH;
|
||||
}
|
||||
} else {
|
||||
if (ARGON2_MIN_AD_LENGTH > context->adlen) {
|
||||
return ARGON2_AD_TOO_SHORT;
|
||||
}
|
||||
if (ARGON2_MAX_AD_LENGTH < context->adlen) {
|
||||
return ARGON2_AD_TOO_LONG;
|
||||
}
|
||||
}
|
||||
|
||||
/* Validate memory cost */
|
||||
if (ARGON2_MIN_MEMORY > context->m_cost) {
|
||||
return ARGON2_MEMORY_TOO_LITTLE;
|
||||
}
|
||||
|
||||
if (ARGON2_MAX_MEMORY < context->m_cost) {
|
||||
return ARGON2_MEMORY_TOO_MUCH;
|
||||
}
|
||||
|
||||
if (context->m_cost < 8 * context->lanes) {
|
||||
return ARGON2_MEMORY_TOO_LITTLE;
|
||||
}
|
||||
|
||||
/* Validate time cost */
|
||||
if (ARGON2_MIN_TIME > context->t_cost) {
|
||||
return ARGON2_TIME_TOO_SMALL;
|
||||
}
|
||||
|
||||
if (ARGON2_MAX_TIME < context->t_cost) {
|
||||
return ARGON2_TIME_TOO_LARGE;
|
||||
}
|
||||
|
||||
/* Validate lanes */
|
||||
if (ARGON2_MIN_LANES > context->lanes) {
|
||||
return ARGON2_LANES_TOO_FEW;
|
||||
}
|
||||
|
||||
if (ARGON2_MAX_LANES < context->lanes) {
|
||||
return ARGON2_LANES_TOO_MANY;
|
||||
}
|
||||
|
||||
/* Validate threads */
|
||||
if (ARGON2_MIN_THREADS > context->threads) {
|
||||
return ARGON2_THREADS_TOO_FEW;
|
||||
}
|
||||
|
||||
if (ARGON2_MAX_THREADS < context->threads) {
|
||||
return ARGON2_THREADS_TOO_MANY;
|
||||
}
|
||||
|
||||
if (NULL != context->allocate_cbk && NULL == context->free_cbk) {
|
||||
return ARGON2_FREE_MEMORY_CBK_NULL;
|
||||
}
|
||||
|
||||
if (NULL == context->allocate_cbk && NULL != context->free_cbk) {
|
||||
return ARGON2_ALLOCATE_MEMORY_CBK_NULL;
|
||||
}
|
||||
|
||||
return ARGON2_OK;
|
||||
}
|
||||
|
||||
void fill_first_blocks(uint8_t *blockhash, const argon2_instance_t *instance) {
|
||||
uint32_t l;
|
||||
/* Make the first and second block in each lane as G(H0||0||i) or
|
||||
G(H0||1||i) */
|
||||
uint8_t blockhash_bytes[ARGON2_BLOCK_SIZE];
|
||||
for (l = 0; l < instance->lanes; ++l) {
|
||||
|
||||
store32(blockhash + ARGON2_PREHASH_DIGEST_LENGTH, 0);
|
||||
store32(blockhash + ARGON2_PREHASH_DIGEST_LENGTH + 4, l);
|
||||
blake2b_long(blockhash_bytes, ARGON2_BLOCK_SIZE, blockhash,
|
||||
ARGON2_PREHASH_SEED_LENGTH);
|
||||
load_block(&instance->memory[l * instance->lane_length + 0],
|
||||
blockhash_bytes);
|
||||
|
||||
store32(blockhash + ARGON2_PREHASH_DIGEST_LENGTH, 1);
|
||||
blake2b_long(blockhash_bytes, ARGON2_BLOCK_SIZE, blockhash,
|
||||
ARGON2_PREHASH_SEED_LENGTH);
|
||||
load_block(&instance->memory[l * instance->lane_length + 1],
|
||||
blockhash_bytes);
|
||||
}
|
||||
clear_internal_memory(blockhash_bytes, ARGON2_BLOCK_SIZE);
|
||||
}
|
||||
|
||||
void initial_hash(uint8_t *blockhash, argon2_context *context,
|
||||
argon2_type type) {
|
||||
blake2b_state BlakeHash;
|
||||
uint8_t value[sizeof(uint32_t)];
|
||||
|
||||
if (NULL == context || NULL == blockhash) {
|
||||
return;
|
||||
}
|
||||
|
||||
blake2b_init(&BlakeHash, ARGON2_PREHASH_DIGEST_LENGTH);
|
||||
|
||||
store32(&value, context->lanes);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
store32(&value, context->outlen);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
store32(&value, context->m_cost);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
store32(&value, context->t_cost);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
// store32(&value, ARGON2_VERSION_NUMBER);
|
||||
store32(&value, context->version);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
store32(&value, (uint32_t)type);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
store32(&value, context->pwdlen);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
if (context->pwd != NULL) {
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)context->pwd,
|
||||
context->pwdlen);
|
||||
|
||||
if (context->flags & ARGON2_FLAG_CLEAR_PASSWORD) {
|
||||
// secure_wipe_memory(context->pwd, context->pwdlen);
|
||||
context->pwdlen = 0;
|
||||
}
|
||||
}
|
||||
|
||||
store32(&value, context->saltlen);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
if (context->salt != NULL) {
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)context->salt,
|
||||
context->saltlen);
|
||||
}
|
||||
|
||||
store32(&value, context->secretlen);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
if (context->secret != NULL) {
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)context->secret,
|
||||
context->secretlen);
|
||||
|
||||
if (context->flags & ARGON2_FLAG_CLEAR_SECRET) {
|
||||
// secure_wipe_memory(context->secret, context->secretlen);
|
||||
context->secretlen = 0;
|
||||
}
|
||||
}
|
||||
|
||||
store32(&value, context->adlen);
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
|
||||
|
||||
if (context->ad != NULL) {
|
||||
blake2b_update(&BlakeHash, (const uint8_t *)context->ad,
|
||||
context->adlen);
|
||||
}
|
||||
|
||||
blake2b_final(&BlakeHash, blockhash, ARGON2_PREHASH_DIGEST_LENGTH);
|
||||
}
|
||||
|
||||
int initialize(argon2_instance_t *instance, argon2_context *context) {
|
||||
uint8_t blockhash[ARGON2_PREHASH_SEED_LENGTH];
|
||||
int result = ARGON2_OK;
|
||||
|
||||
if (instance == NULL || context == NULL)
|
||||
return ARGON2_INCORRECT_PARAMETER;
|
||||
instance->context_ptr = context;
|
||||
|
||||
/* 1. Memory allocation */
|
||||
result = allocate_memory(context, (uint8_t **)&(instance->memory),
|
||||
instance->memory_blocks, sizeof(block));
|
||||
if (result != ARGON2_OK) {
|
||||
return result;
|
||||
}
|
||||
|
||||
/* 2. Initial hashing */
|
||||
/* H_0 + 8 extra bytes to produce the first blocks */
|
||||
/* uint8_t blockhash[ARGON2_PREHASH_SEED_LENGTH]; */
|
||||
/* Hashing all inputs */
|
||||
initial_hash(blockhash, context, instance->type);
|
||||
/* Zeroing 8 extra bytes */
|
||||
clear_internal_memory(blockhash + ARGON2_PREHASH_DIGEST_LENGTH,
|
||||
ARGON2_PREHASH_SEED_LENGTH -
|
||||
ARGON2_PREHASH_DIGEST_LENGTH);
|
||||
|
||||
#ifdef GENKAT
|
||||
initial_kat(blockhash, context, instance->type);
|
||||
#endif
|
||||
|
||||
/* 3. Creating first blocks, we always have at least two blocks in a slice
|
||||
*/
|
||||
fill_first_blocks(blockhash, instance);
|
||||
/* Clearing the hash */
|
||||
clear_internal_memory(blockhash, ARGON2_PREHASH_SEED_LENGTH);
|
||||
|
||||
return ARGON2_OK;
|
||||
}
|
||||
228
algo/argon2/argon2d/argon2d/core.h
Normal file
228
algo/argon2/argon2d/argon2d/core.h
Normal file
@@ -0,0 +1,228 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#ifndef ARGON2_CORE_H
|
||||
#define ARGON2_CORE_H
|
||||
|
||||
#include "argon2.h"
|
||||
|
||||
#define CONST_CAST(x) (x)(uintptr_t)
|
||||
|
||||
/**********************Argon2 internal constants*******************************/
|
||||
|
||||
enum argon2_core_constants {
|
||||
/* Memory block size in bytes */
|
||||
ARGON2_BLOCK_SIZE = 1024,
|
||||
ARGON2_QWORDS_IN_BLOCK = ARGON2_BLOCK_SIZE / 8,
|
||||
ARGON2_OWORDS_IN_BLOCK = ARGON2_BLOCK_SIZE / 16,
|
||||
ARGON2_HWORDS_IN_BLOCK = ARGON2_BLOCK_SIZE / 32,
|
||||
ARGON2_512BIT_WORDS_IN_BLOCK = ARGON2_BLOCK_SIZE / 64,
|
||||
|
||||
/* Number of pseudo-random values generated by one call to Blake in Argon2i
|
||||
to
|
||||
generate reference block positions */
|
||||
ARGON2_ADDRESSES_IN_BLOCK = 128,
|
||||
|
||||
/* Pre-hashing digest length and its extension*/
|
||||
ARGON2_PREHASH_DIGEST_LENGTH = 64,
|
||||
ARGON2_PREHASH_SEED_LENGTH = 72
|
||||
};
|
||||
|
||||
/*************************Argon2 internal data types***********************/
|
||||
|
||||
/*
|
||||
* Structure for the (1KB) memory block implemented as 128 64-bit words.
|
||||
* Memory blocks can be copied, XORed. Internal words can be accessed by [] (no
|
||||
* bounds checking).
|
||||
*/
|
||||
typedef struct block_ { uint64_t v[ARGON2_QWORDS_IN_BLOCK]; } block;
|
||||
|
||||
/*****************Functions that work with the block******************/
|
||||
|
||||
/* Initialize each byte of the block with @in */
|
||||
void init_block_value(block *b, uint8_t in);
|
||||
|
||||
/* Copy block @src to block @dst */
|
||||
void copy_block(block *dst, const block *src);
|
||||
|
||||
/* XOR @src onto @dst bytewise */
|
||||
void xor_block(block *dst, const block *src);
|
||||
|
||||
/*
|
||||
* Argon2 instance: memory pointer, number of passes, amount of memory, type,
|
||||
* and derived values.
|
||||
* Used to evaluate the number and location of blocks to construct in each
|
||||
* thread
|
||||
*/
|
||||
typedef struct Argon2_instance_t {
|
||||
block *memory; /* Memory pointer */
|
||||
uint32_t version;
|
||||
uint32_t passes; /* Number of passes */
|
||||
uint32_t memory_blocks; /* Number of blocks in memory */
|
||||
uint32_t segment_length;
|
||||
uint32_t lane_length;
|
||||
uint32_t lanes;
|
||||
uint32_t threads;
|
||||
argon2_type type;
|
||||
int print_internals; /* whether to print the memory blocks */
|
||||
argon2_context *context_ptr; /* points back to original context */
|
||||
} argon2_instance_t;
|
||||
|
||||
/*
|
||||
* Argon2 position: where we construct the block right now. Used to distribute
|
||||
* work between threads.
|
||||
*/
|
||||
typedef struct Argon2_position_t {
|
||||
uint32_t pass;
|
||||
uint32_t lane;
|
||||
uint8_t slice;
|
||||
uint32_t index;
|
||||
} argon2_position_t;
|
||||
|
||||
/*Struct that holds the inputs for thread handling FillSegment*/
|
||||
typedef struct Argon2_thread_data {
|
||||
argon2_instance_t *instance_ptr;
|
||||
argon2_position_t pos;
|
||||
} argon2_thread_data;
|
||||
|
||||
/*************************Argon2 core functions********************************/
|
||||
|
||||
/* Allocates memory to the given pointer, uses the appropriate allocator as
|
||||
* specified in the context. Total allocated memory is num*size.
|
||||
* @param context argon2_context which specifies the allocator
|
||||
* @param memory pointer to the pointer to the memory
|
||||
* @param size the size in bytes for each element to be allocated
|
||||
* @param num the number of elements to be allocated
|
||||
* @return ARGON2_OK if @memory is a valid pointer and memory is allocated
|
||||
*/
|
||||
int allocate_memory(const argon2_context *context, uint8_t **memory,
|
||||
size_t num, size_t size);
|
||||
|
||||
/*
|
||||
* Frees memory at the given pointer, uses the appropriate deallocator as
|
||||
* specified in the context. Also cleans the memory using clear_internal_memory.
|
||||
* @param context argon2_context which specifies the deallocator
|
||||
* @param memory pointer to buffer to be freed
|
||||
* @param size the size in bytes for each element to be deallocated
|
||||
* @param num the number of elements to be deallocated
|
||||
*/
|
||||
void free_memory(const argon2_context *context, uint8_t *memory,
|
||||
size_t num, size_t size);
|
||||
|
||||
/* Function that securely cleans the memory. This ignores any flags set
|
||||
* regarding clearing memory. Usually one just calls clear_internal_memory.
|
||||
* @param mem Pointer to the memory
|
||||
* @param s Memory size in bytes
|
||||
*/
|
||||
void secure_wipe_memory(void *v, size_t n);
|
||||
|
||||
/* Function that securely clears the memory if FLAG_clear_internal_memory is
|
||||
* set. If the flag isn't set, this function does nothing.
|
||||
* @param mem Pointer to the memory
|
||||
* @param s Memory size in bytes
|
||||
*/
|
||||
void clear_internal_memory(void *v, size_t n);
|
||||
|
||||
/*
|
||||
* Computes absolute position of reference block in the lane following a skewed
|
||||
* distribution and using a pseudo-random value as input
|
||||
* @param instance Pointer to the current instance
|
||||
* @param position Pointer to the current position
|
||||
* @param pseudo_rand 32-bit pseudo-random value used to determine the position
|
||||
* @param same_lane Indicates if the block will be taken from the current lane.
|
||||
* If so we can reference the current segment
|
||||
* @pre All pointers must be valid
|
||||
*/
|
||||
uint32_t index_alpha(const argon2_instance_t *instance,
|
||||
const argon2_position_t *position, uint32_t pseudo_rand,
|
||||
int same_lane);
|
||||
|
||||
/*
|
||||
* Function that validates all inputs against predefined restrictions and return
|
||||
* an error code
|
||||
* @param context Pointer to current Argon2 context
|
||||
* @return ARGON2_OK if everything is all right, otherwise one of error codes
|
||||
* (all defined in <argon2.h>
|
||||
*/
|
||||
int validate_inputs(const argon2_context *context);
|
||||
|
||||
/*
|
||||
* Hashes all the inputs into @a blockhash[PREHASH_DIGEST_LENGTH], clears
|
||||
* password and secret if needed
|
||||
* @param context Pointer to the Argon2 internal structure containing memory
|
||||
* pointer, and parameters for time and space requirements.
|
||||
* @param blockhash Buffer for pre-hashing digest
|
||||
* @param type Argon2 type
|
||||
* @pre @a blockhash must have at least @a PREHASH_DIGEST_LENGTH bytes
|
||||
* allocated
|
||||
*/
|
||||
void initial_hash(uint8_t *blockhash, argon2_context *context,
|
||||
argon2_type type);
|
||||
|
||||
/*
|
||||
* Function creates first 2 blocks per lane
|
||||
* @param instance Pointer to the current instance
|
||||
* @param blockhash Pointer to the pre-hashing digest
|
||||
* @pre blockhash must point to @a PREHASH_SEED_LENGTH allocated values
|
||||
*/
|
||||
void fill_first_blocks(uint8_t *blockhash, const argon2_instance_t *instance);
|
||||
|
||||
/*
|
||||
* Function allocates memory, hashes the inputs with Blake, and creates first
|
||||
* two blocks. Returns the pointer to the main memory with 2 blocks per lane
|
||||
* initialized
|
||||
* @param context Pointer to the Argon2 internal structure containing memory
|
||||
* pointer, and parameters for time and space requirements.
|
||||
* @param instance Current Argon2 instance
|
||||
* @return Zero if successful, -1 if memory failed to allocate. @context->state
|
||||
* will be modified if successful.
|
||||
*/
|
||||
int initialize(argon2_instance_t *instance, argon2_context *context);
|
||||
|
||||
/*
|
||||
* XORing the last block of each lane, hashing it, making the tag. Deallocates
|
||||
* the memory.
|
||||
* @param context Pointer to current Argon2 context (use only the out parameters
|
||||
* from it)
|
||||
* @param instance Pointer to current instance of Argon2
|
||||
* @pre instance->state must point to necessary amount of memory
|
||||
* @pre context->out must point to outlen bytes of memory
|
||||
* @pre if context->free_cbk is not NULL, it should point to a function that
|
||||
* deallocates memory
|
||||
*/
|
||||
void finalize(const argon2_context *context, argon2_instance_t *instance);
|
||||
|
||||
/*
|
||||
* Function that fills the segment using previous segments also from other
|
||||
* threads
|
||||
* @param context current context
|
||||
* @param instance Pointer to the current instance
|
||||
* @param position Current position
|
||||
* @pre all block pointers must be valid
|
||||
*/
|
||||
void fill_segment(const argon2_instance_t *instance,
|
||||
argon2_position_t position);
|
||||
|
||||
/*
|
||||
* Function that fills the entire memory t_cost times based on the first two
|
||||
* blocks in each lane
|
||||
* @param instance Pointer to the current instance
|
||||
* @return ARGON2_OK if successful, @context->state
|
||||
*/
|
||||
int fill_memory_blocks(argon2_instance_t *instance);
|
||||
|
||||
#endif
|
||||
463
algo/argon2/argon2d/argon2d/encoding.c
Normal file
463
algo/argon2/argon2d/argon2d/encoding.c
Normal file
@@ -0,0 +1,463 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
#include <limits.h>
|
||||
#include "encoding.h"
|
||||
#include "core.h"
|
||||
|
||||
/*
|
||||
* Example code for a decoder and encoder of "hash strings", with Argon2
|
||||
* parameters.
|
||||
*
|
||||
* This code comprises three sections:
|
||||
*
|
||||
* -- The first section contains generic Base64 encoding and decoding
|
||||
* functions. It is conceptually applicable to any hash function
|
||||
* implementation that uses Base64 to encode and decode parameters,
|
||||
* salts and outputs. It could be made into a library, provided that
|
||||
* the relevant functions are made public (non-static) and be given
|
||||
* reasonable names to avoid collisions with other functions.
|
||||
*
|
||||
* -- The second section is specific to Argon2. It encodes and decodes
|
||||
* the parameters, salts and outputs. It does not compute the hash
|
||||
* itself.
|
||||
*
|
||||
* The code was originally written by Thomas Pornin <pornin@bolet.org>,
|
||||
* to whom comments and remarks may be sent. It is released under what
|
||||
* should amount to Public Domain or its closest equivalent; the
|
||||
* following mantra is supposed to incarnate that fact with all the
|
||||
* proper legal rituals:
|
||||
*
|
||||
* ---------------------------------------------------------------------
|
||||
* This file is provided under the terms of Creative Commons CC0 1.0
|
||||
* Public Domain Dedication. To the extent possible under law, the
|
||||
* author (Thomas Pornin) has waived all copyright and related or
|
||||
* neighboring rights to this file. This work is published from: Canada.
|
||||
* ---------------------------------------------------------------------
|
||||
*
|
||||
* Copyright (c) 2015 Thomas Pornin
|
||||
*/
|
||||
|
||||
/* ==================================================================== */
|
||||
/*
|
||||
* Common code; could be shared between different hash functions.
|
||||
*
|
||||
* Note: the Base64 functions below assume that uppercase letters (resp.
|
||||
* lowercase letters) have consecutive numerical codes, that fit on 8
|
||||
* bits. All modern systems use ASCII-compatible charsets, where these
|
||||
* properties are true. If you are stuck with a dinosaur of a system
|
||||
* that still defaults to EBCDIC then you already have much bigger
|
||||
* interoperability issues to deal with.
|
||||
*/
|
||||
|
||||
/*
|
||||
* Some macros for constant-time comparisons. These work over values in
|
||||
* the 0..255 range. Returned value is 0x00 on "false", 0xFF on "true".
|
||||
*/
|
||||
#define EQ(x, y) ((((0U - ((unsigned)(x) ^ (unsigned)(y))) >> 8) & 0xFF) ^ 0xFF)
|
||||
#define GT(x, y) ((((unsigned)(y) - (unsigned)(x)) >> 8) & 0xFF)
|
||||
#define GE(x, y) (GT(y, x) ^ 0xFF)
|
||||
#define LT(x, y) GT(y, x)
|
||||
#define LE(x, y) GE(y, x)
|
||||
|
||||
/*
|
||||
* Convert value x (0..63) to corresponding Base64 character.
|
||||
*/
|
||||
static int b64_byte_to_char(unsigned x) {
|
||||
return (LT(x, 26) & (x + 'A')) |
|
||||
(GE(x, 26) & LT(x, 52) & (x + ('a' - 26))) |
|
||||
(GE(x, 52) & LT(x, 62) & (x + ('0' - 52))) | (EQ(x, 62) & '+') |
|
||||
(EQ(x, 63) & '/');
|
||||
}
|
||||
|
||||
/*
|
||||
* Convert character c to the corresponding 6-bit value. If character c
|
||||
* is not a Base64 character, then 0xFF (255) is returned.
|
||||
*/
|
||||
static unsigned b64_char_to_byte(int c) {
|
||||
unsigned x;
|
||||
|
||||
x = (GE(c, 'A') & LE(c, 'Z') & (c - 'A')) |
|
||||
(GE(c, 'a') & LE(c, 'z') & (c - ('a' - 26))) |
|
||||
(GE(c, '0') & LE(c, '9') & (c - ('0' - 52))) | (EQ(c, '+') & 62) |
|
||||
(EQ(c, '/') & 63);
|
||||
return x | (EQ(x, 0) & (EQ(c, 'A') ^ 0xFF));
|
||||
}
|
||||
|
||||
/*
|
||||
* Convert some bytes to Base64. 'dst_len' is the length (in characters)
|
||||
* of the output buffer 'dst'; if that buffer is not large enough to
|
||||
* receive the result (including the terminating 0), then (size_t)-1
|
||||
* is returned. Otherwise, the zero-terminated Base64 string is written
|
||||
* in the buffer, and the output length (counted WITHOUT the terminating
|
||||
* zero) is returned.
|
||||
*/
|
||||
static size_t to_base64(char *dst, size_t dst_len, const void *src,
|
||||
size_t src_len) {
|
||||
size_t olen;
|
||||
const unsigned char *buf;
|
||||
unsigned acc, acc_len;
|
||||
|
||||
olen = (src_len / 3) << 2;
|
||||
switch (src_len % 3) {
|
||||
case 2:
|
||||
olen++;
|
||||
/* fall through */
|
||||
case 1:
|
||||
olen += 2;
|
||||
break;
|
||||
}
|
||||
if (dst_len <= olen) {
|
||||
return (size_t)-1;
|
||||
}
|
||||
acc = 0;
|
||||
acc_len = 0;
|
||||
buf = (const unsigned char *)src;
|
||||
while (src_len-- > 0) {
|
||||
acc = (acc << 8) + (*buf++);
|
||||
acc_len += 8;
|
||||
while (acc_len >= 6) {
|
||||
acc_len -= 6;
|
||||
*dst++ = (char)b64_byte_to_char((acc >> acc_len) & 0x3F);
|
||||
}
|
||||
}
|
||||
if (acc_len > 0) {
|
||||
*dst++ = (char)b64_byte_to_char((acc << (6 - acc_len)) & 0x3F);
|
||||
}
|
||||
*dst++ = 0;
|
||||
return olen;
|
||||
}
|
||||
|
||||
/*
|
||||
* Decode Base64 chars into bytes. The '*dst_len' value must initially
|
||||
* contain the length of the output buffer '*dst'; when the decoding
|
||||
* ends, the actual number of decoded bytes is written back in
|
||||
* '*dst_len'.
|
||||
*
|
||||
* Decoding stops when a non-Base64 character is encountered, or when
|
||||
* the output buffer capacity is exceeded. If an error occurred (output
|
||||
* buffer is too small, invalid last characters leading to unprocessed
|
||||
* buffered bits), then NULL is returned; otherwise, the returned value
|
||||
* points to the first non-Base64 character in the source stream, which
|
||||
* may be the terminating zero.
|
||||
*/
|
||||
static const char *from_base64(void *dst, size_t *dst_len, const char *src) {
|
||||
size_t len;
|
||||
unsigned char *buf;
|
||||
unsigned acc, acc_len;
|
||||
|
||||
buf = (unsigned char *)dst;
|
||||
len = 0;
|
||||
acc = 0;
|
||||
acc_len = 0;
|
||||
for (;;) {
|
||||
unsigned d;
|
||||
|
||||
d = b64_char_to_byte(*src);
|
||||
if (d == 0xFF) {
|
||||
break;
|
||||
}
|
||||
src++;
|
||||
acc = (acc << 6) + d;
|
||||
acc_len += 6;
|
||||
if (acc_len >= 8) {
|
||||
acc_len -= 8;
|
||||
if ((len++) >= *dst_len) {
|
||||
return NULL;
|
||||
}
|
||||
*buf++ = (acc >> acc_len) & 0xFF;
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
* If the input length is equal to 1 modulo 4 (which is
|
||||
* invalid), then there will remain 6 unprocessed bits;
|
||||
* otherwise, only 0, 2 or 4 bits are buffered. The buffered
|
||||
* bits must also all be zero.
|
||||
*/
|
||||
if (acc_len > 4 || (acc & (((unsigned)1 << acc_len) - 1)) != 0) {
|
||||
return NULL;
|
||||
}
|
||||
*dst_len = len;
|
||||
return src;
|
||||
}
|
||||
|
||||
/*
|
||||
* Decode decimal integer from 'str'; the value is written in '*v'.
|
||||
* Returned value is a pointer to the next non-decimal character in the
|
||||
* string. If there is no digit at all, or the value encoding is not
|
||||
* minimal (extra leading zeros), or the value does not fit in an
|
||||
* 'unsigned long', then NULL is returned.
|
||||
*/
|
||||
static const char *decode_decimal(const char *str, unsigned long *v) {
|
||||
const char *orig;
|
||||
unsigned long acc;
|
||||
|
||||
acc = 0;
|
||||
for (orig = str;; str++) {
|
||||
int c;
|
||||
|
||||
c = *str;
|
||||
if (c < '0' || c > '9') {
|
||||
break;
|
||||
}
|
||||
c -= '0';
|
||||
if (acc > (ULONG_MAX / 10)) {
|
||||
return NULL;
|
||||
}
|
||||
acc *= 10;
|
||||
if ((unsigned long)c > (ULONG_MAX - acc)) {
|
||||
return NULL;
|
||||
}
|
||||
acc += (unsigned long)c;
|
||||
}
|
||||
if (str == orig || (*orig == '0' && str != (orig + 1))) {
|
||||
return NULL;
|
||||
}
|
||||
*v = acc;
|
||||
return str;
|
||||
}
|
||||
|
||||
/* ==================================================================== */
|
||||
/*
|
||||
* Code specific to Argon2.
|
||||
*
|
||||
* The code below applies the following format:
|
||||
*
|
||||
* $argon2<T>[$v=<num>]$m=<num>,t=<num>,p=<num>$<bin>$<bin>
|
||||
*
|
||||
* where <T> is either 'd', 'id', or 'i', <num> is a decimal integer (positive,
|
||||
* fits in an 'unsigned long'), and <bin> is Base64-encoded data (no '=' padding
|
||||
* characters, no newline or whitespace).
|
||||
*
|
||||
* The last two binary chunks (encoded in Base64) are, in that order,
|
||||
* the salt and the output. Both are required. The binary salt length and the
|
||||
* output length must be in the allowed ranges defined in argon2.h.
|
||||
*
|
||||
* The ctx struct must contain buffers large enough to hold the salt and pwd
|
||||
* when it is fed into decode_string.
|
||||
*/
|
||||
|
||||
int decode_string(argon2_context *ctx, const char *str, argon2_type type) {
|
||||
|
||||
/* check for prefix */
|
||||
#define CC(prefix) \
|
||||
do { \
|
||||
size_t cc_len = strlen(prefix); \
|
||||
if (strncmp(str, prefix, cc_len) != 0) { \
|
||||
return ARGON2_DECODING_FAIL; \
|
||||
} \
|
||||
str += cc_len; \
|
||||
} while ((void)0, 0)
|
||||
|
||||
/* optional prefix checking with supplied code */
|
||||
#define CC_opt(prefix, code) \
|
||||
do { \
|
||||
size_t cc_len = strlen(prefix); \
|
||||
if (strncmp(str, prefix, cc_len) == 0) { \
|
||||
str += cc_len; \
|
||||
{ code; } \
|
||||
} \
|
||||
} while ((void)0, 0)
|
||||
|
||||
/* Decoding prefix into decimal */
|
||||
#define DECIMAL(x) \
|
||||
do { \
|
||||
unsigned long dec_x; \
|
||||
str = decode_decimal(str, &dec_x); \
|
||||
if (str == NULL) { \
|
||||
return ARGON2_DECODING_FAIL; \
|
||||
} \
|
||||
(x) = dec_x; \
|
||||
} while ((void)0, 0)
|
||||
|
||||
|
||||
/* Decoding prefix into uint32_t decimal */
|
||||
#define DECIMAL_U32(x) \
|
||||
do { \
|
||||
unsigned long dec_x; \
|
||||
str = decode_decimal(str, &dec_x); \
|
||||
if (str == NULL || dec_x > UINT32_MAX) { \
|
||||
return ARGON2_DECODING_FAIL; \
|
||||
} \
|
||||
(x) = (uint32_t)dec_x; \
|
||||
} while ((void)0, 0)
|
||||
|
||||
|
||||
/* Decoding base64 into a binary buffer */
|
||||
#define BIN(buf, max_len, len) \
|
||||
do { \
|
||||
size_t bin_len = (max_len); \
|
||||
str = from_base64(buf, &bin_len, str); \
|
||||
if (str == NULL || bin_len > UINT32_MAX) { \
|
||||
return ARGON2_DECODING_FAIL; \
|
||||
} \
|
||||
(len) = (uint32_t)bin_len; \
|
||||
} while ((void)0, 0)
|
||||
|
||||
size_t maxsaltlen = ctx->saltlen;
|
||||
size_t maxoutlen = ctx->outlen;
|
||||
int validation_result;
|
||||
const char* type_string;
|
||||
|
||||
/* We should start with the argon2_type we are using */
|
||||
type_string = argon2_type2string(type, 0);
|
||||
if (!type_string) {
|
||||
return ARGON2_INCORRECT_TYPE;
|
||||
}
|
||||
|
||||
CC("$");
|
||||
CC(type_string);
|
||||
|
||||
/* Reading the version number if the default is suppressed */
|
||||
ctx->version = ARGON2_VERSION_10;
|
||||
CC_opt("$v=", DECIMAL_U32(ctx->version));
|
||||
|
||||
CC("$m=");
|
||||
DECIMAL_U32(ctx->m_cost);
|
||||
CC(",t=");
|
||||
DECIMAL_U32(ctx->t_cost);
|
||||
CC(",p=");
|
||||
DECIMAL_U32(ctx->lanes);
|
||||
ctx->threads = ctx->lanes;
|
||||
|
||||
CC("$");
|
||||
BIN(ctx->salt, maxsaltlen, ctx->saltlen);
|
||||
CC("$");
|
||||
BIN(ctx->out, maxoutlen, ctx->outlen);
|
||||
|
||||
/* The rest of the fields get the default values */
|
||||
ctx->secret = NULL;
|
||||
ctx->secretlen = 0;
|
||||
ctx->ad = NULL;
|
||||
ctx->adlen = 0;
|
||||
ctx->allocate_cbk = NULL;
|
||||
ctx->free_cbk = NULL;
|
||||
ctx->flags = ARGON2_DEFAULT_FLAGS;
|
||||
|
||||
/* On return, must have valid context */
|
||||
validation_result = validate_inputs(ctx);
|
||||
if (validation_result != ARGON2_OK) {
|
||||
return validation_result;
|
||||
}
|
||||
|
||||
/* Can't have any additional characters */
|
||||
if (*str == 0) {
|
||||
return ARGON2_OK;
|
||||
} else {
|
||||
return ARGON2_DECODING_FAIL;
|
||||
}
|
||||
#undef CC
|
||||
#undef CC_opt
|
||||
#undef DECIMAL
|
||||
#undef BIN
|
||||
}
|
||||
|
||||
int encode_string(char *dst, size_t dst_len, argon2_context *ctx,
|
||||
argon2_type type) {
|
||||
#define SS(str) \
|
||||
do { \
|
||||
size_t pp_len = strlen(str); \
|
||||
if (pp_len >= dst_len) { \
|
||||
return ARGON2_ENCODING_FAIL; \
|
||||
} \
|
||||
memcpy(dst, str, pp_len + 1); \
|
||||
dst += pp_len; \
|
||||
dst_len -= pp_len; \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define SX(x) \
|
||||
do { \
|
||||
char tmp[30]; \
|
||||
sprintf(tmp, "%lu", (unsigned long)(x)); \
|
||||
SS(tmp); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define SB(buf, len) \
|
||||
do { \
|
||||
size_t sb_len = to_base64(dst, dst_len, buf, len); \
|
||||
if (sb_len == (size_t)-1) { \
|
||||
return ARGON2_ENCODING_FAIL; \
|
||||
} \
|
||||
dst += sb_len; \
|
||||
dst_len -= sb_len; \
|
||||
} while ((void)0, 0)
|
||||
|
||||
const char* type_string = argon2_type2string(type, 0);
|
||||
int validation_result = validate_inputs(ctx);
|
||||
|
||||
if (!type_string) {
|
||||
return ARGON2_ENCODING_FAIL;
|
||||
}
|
||||
|
||||
if (validation_result != ARGON2_OK) {
|
||||
return validation_result;
|
||||
}
|
||||
|
||||
|
||||
SS("$");
|
||||
SS(type_string);
|
||||
|
||||
SS("$v=");
|
||||
SX(ctx->version);
|
||||
|
||||
SS("$m=");
|
||||
SX(ctx->m_cost);
|
||||
SS(",t=");
|
||||
SX(ctx->t_cost);
|
||||
SS(",p=");
|
||||
SX(ctx->lanes);
|
||||
|
||||
SS("$");
|
||||
SB(ctx->salt, ctx->saltlen);
|
||||
|
||||
SS("$");
|
||||
SB(ctx->out, ctx->outlen);
|
||||
return ARGON2_OK;
|
||||
|
||||
#undef SS
|
||||
#undef SX
|
||||
#undef SB
|
||||
}
|
||||
|
||||
size_t b64len(uint32_t len) {
|
||||
size_t olen = ((size_t)len / 3) << 2;
|
||||
|
||||
switch (len % 3) {
|
||||
case 2:
|
||||
olen++;
|
||||
/* fall through */
|
||||
case 1:
|
||||
olen += 2;
|
||||
break;
|
||||
}
|
||||
|
||||
return olen;
|
||||
}
|
||||
|
||||
size_t numlen(uint32_t num) {
|
||||
size_t len = 1;
|
||||
while (num >= 10) {
|
||||
++len;
|
||||
num = num / 10;
|
||||
}
|
||||
return len;
|
||||
}
|
||||
|
||||
57
algo/argon2/argon2d/argon2d/encoding.h
Normal file
57
algo/argon2/argon2d/argon2d/encoding.h
Normal file
@@ -0,0 +1,57 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#ifndef ENCODING_H
|
||||
#define ENCODING_H
|
||||
#include "argon2.h"
|
||||
|
||||
#define ARGON2_MAX_DECODED_LANES UINT32_C(255)
|
||||
#define ARGON2_MIN_DECODED_SALT_LEN UINT32_C(8)
|
||||
#define ARGON2_MIN_DECODED_OUT_LEN UINT32_C(12)
|
||||
|
||||
/*
|
||||
* encode an Argon2 hash string into the provided buffer. 'dst_len'
|
||||
* contains the size, in characters, of the 'dst' buffer; if 'dst_len'
|
||||
* is less than the number of required characters (including the
|
||||
* terminating 0), then this function returns ARGON2_ENCODING_ERROR.
|
||||
*
|
||||
* on success, ARGON2_OK is returned.
|
||||
*/
|
||||
int encode_string(char *dst, size_t dst_len, argon2_context *ctx,
|
||||
argon2_type type);
|
||||
|
||||
/*
|
||||
* Decodes an Argon2 hash string into the provided structure 'ctx'.
|
||||
* The only fields that must be set prior to this call are ctx.saltlen and
|
||||
* ctx.outlen (which must be the maximal salt and out length values that are
|
||||
* allowed), ctx.salt and ctx.out (which must be buffers of the specified
|
||||
* length), and ctx.pwd and ctx.pwdlen which must hold a valid password.
|
||||
*
|
||||
* Invalid input string causes an error. On success, the ctx is valid and all
|
||||
* fields have been initialized.
|
||||
*
|
||||
* Returned value is ARGON2_OK on success, other ARGON2_ codes on error.
|
||||
*/
|
||||
int decode_string(argon2_context *ctx, const char *str, argon2_type type);
|
||||
|
||||
/* Returns the length of the encoded byte stream with length len */
|
||||
size_t b64len(uint32_t len);
|
||||
|
||||
/* Returns the length of the encoded number num */
|
||||
size_t numlen(uint32_t num);
|
||||
|
||||
#endif
|
||||
359
algo/argon2/argon2d/argon2d/opt.c
Normal file
359
algo/argon2/argon2d/argon2d/opt.c
Normal file
@@ -0,0 +1,359 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#include <stdint.h>
|
||||
#include <string.h>
|
||||
#include <stdlib.h>
|
||||
|
||||
#include "argon2.h"
|
||||
#include "core.h"
|
||||
|
||||
#include "../blake2/blake2.h"
|
||||
#include "../blake2/blamka-round-opt.h"
|
||||
|
||||
/*
|
||||
* Function fills a new memory block and optionally XORs the old block over the new one.
|
||||
* Memory must be initialized.
|
||||
* @param state Pointer to the just produced block. Content will be updated(!)
|
||||
* @param ref_block Pointer to the reference block
|
||||
* @param next_block Pointer to the block to be XORed over. May coincide with @ref_block
|
||||
* @param with_xor Whether to XOR into the new block (1) or just overwrite (0)
|
||||
* @pre all block pointers must be valid
|
||||
*/
|
||||
|
||||
#if defined(__AVX512F__)
|
||||
|
||||
static void fill_block(__m512i *state, const block *ref_block,
|
||||
block *next_block, int with_xor) {
|
||||
__m512i block_XY[ARGON2_512BIT_WORDS_IN_BLOCK];
|
||||
unsigned int i;
|
||||
|
||||
if (with_xor) {
|
||||
for (i = 0; i < ARGON2_512BIT_WORDS_IN_BLOCK; i++) {
|
||||
state[i] = _mm512_xor_si512(
|
||||
state[i], _mm512_loadu_si512((const __m512i *)ref_block->v + i));
|
||||
block_XY[i] = _mm512_xor_si512(
|
||||
state[i], _mm512_loadu_si512((const __m512i *)next_block->v + i));
|
||||
}
|
||||
} else {
|
||||
for (i = 0; i < ARGON2_512BIT_WORDS_IN_BLOCK; i++) {
|
||||
block_XY[i] = state[i] = _mm512_xor_si512(
|
||||
state[i], _mm512_loadu_si512((const __m512i *)ref_block->v + i));
|
||||
}
|
||||
}
|
||||
|
||||
BLAKE2_ROUND_1( state[ 0], state[ 1], state[ 2], state[ 3],
|
||||
state[ 4], state[ 5], state[ 6], state[ 7] );
|
||||
BLAKE2_ROUND_1( state[ 8], state[ 9], state[10], state[11],
|
||||
state[12], state[13], state[14], state[15] );
|
||||
|
||||
BLAKE2_ROUND_2( state[ 0], state[ 2], state[ 4], state[ 6],
|
||||
state[ 8], state[10], state[12], state[14] );
|
||||
BLAKE2_ROUND_2( state[ 1], state[ 3], state[ 5], state[ 7],
|
||||
state[ 9], state[11], state[13], state[15] );
|
||||
|
||||
/*
|
||||
for (i = 0; i < 2; ++i) {
|
||||
BLAKE2_ROUND_1(
|
||||
state[8 * i + 0], state[8 * i + 1], state[8 * i + 2], state[8 * i + 3],
|
||||
state[8 * i + 4], state[8 * i + 5], state[8 * i + 6], state[8 * i + 7]);
|
||||
}
|
||||
|
||||
for (i = 0; i < 2; ++i) {
|
||||
BLAKE2_ROUND_2(
|
||||
state[2 * 0 + i], state[2 * 1 + i], state[2 * 2 + i], state[2 * 3 + i],
|
||||
state[2 * 4 + i], state[2 * 5 + i], state[2 * 6 + i], state[2 * 7 + i]);
|
||||
}
|
||||
*/
|
||||
|
||||
for (i = 0; i < ARGON2_512BIT_WORDS_IN_BLOCK; i++) {
|
||||
state[i] = _mm512_xor_si512(state[i], block_XY[i]);
|
||||
_mm512_storeu_si512((__m512i *)next_block->v + i, state[i]);
|
||||
}
|
||||
}
|
||||
|
||||
#elif defined(__AVX2__)
|
||||
|
||||
static void fill_block(__m256i *state, const block *ref_block,
|
||||
block *next_block, int with_xor) {
|
||||
__m256i block_XY[ARGON2_HWORDS_IN_BLOCK];
|
||||
unsigned int i;
|
||||
|
||||
if (with_xor) {
|
||||
for (i = 0; i < ARGON2_HWORDS_IN_BLOCK; i++) {
|
||||
state[i] = _mm256_xor_si256(
|
||||
state[i], _mm256_loadu_si256((const __m256i *)ref_block->v + i));
|
||||
block_XY[i] = _mm256_xor_si256(
|
||||
state[i], _mm256_loadu_si256((const __m256i *)next_block->v + i));
|
||||
}
|
||||
} else {
|
||||
for (i = 0; i < ARGON2_HWORDS_IN_BLOCK; i++) {
|
||||
block_XY[i] = state[i] = _mm256_xor_si256(
|
||||
state[i], _mm256_loadu_si256((const __m256i *)ref_block->v + i));
|
||||
}
|
||||
}
|
||||
|
||||
BLAKE2_ROUND_1( state[ 0], state[ 4], state[ 1], state[ 5],
|
||||
state[ 2], state[ 6], state[ 3], state[ 7] );
|
||||
BLAKE2_ROUND_1( state[ 8], state[12], state[ 9], state[13],
|
||||
state[10], state[14], state[11], state[15] );
|
||||
BLAKE2_ROUND_1( state[16], state[20], state[17], state[21],
|
||||
state[18], state[22], state[19], state[23] );
|
||||
BLAKE2_ROUND_1( state[24], state[28], state[25], state[29],
|
||||
state[26], state[30], state[27], state[31] );
|
||||
|
||||
BLAKE2_ROUND_2( state[ 0], state[ 4], state[ 8], state[12],
|
||||
state[16], state[20], state[24], state[28] );
|
||||
BLAKE2_ROUND_2( state[ 1], state[ 5], state[ 9], state[13],
|
||||
state[17], state[21], state[25], state[29] );
|
||||
BLAKE2_ROUND_2( state[ 2], state[ 6], state[10], state[14],
|
||||
state[18], state[22], state[26], state[30] );
|
||||
BLAKE2_ROUND_2( state[ 3], state[ 7], state[11], state[15],
|
||||
state[19], state[23], state[27], state[31] );
|
||||
|
||||
/*
|
||||
for (i = 0; i < 4; ++i) {
|
||||
BLAKE2_ROUND_1(state[8 * i + 0], state[8 * i + 4], state[8 * i + 1], state[8 * i + 5],
|
||||
state[8 * i + 2], state[8 * i + 6], state[8 * i + 3], state[8 * i + 7]);
|
||||
}
|
||||
|
||||
for (i = 0; i < 4; ++i) {
|
||||
BLAKE2_ROUND_2(state[ 0 + i], state[ 4 + i], state[ 8 + i], state[12 + i],
|
||||
state[16 + i], state[20 + i], state[24 + i], state[28 + i]);
|
||||
}
|
||||
*/
|
||||
|
||||
for (i = 0; i < ARGON2_HWORDS_IN_BLOCK; i++) {
|
||||
state[i] = _mm256_xor_si256(state[i], block_XY[i]);
|
||||
_mm256_storeu_si256((__m256i *)next_block->v + i, state[i]);
|
||||
}
|
||||
}
|
||||
|
||||
#else // SSE2
|
||||
|
||||
static void fill_block(__m128i *state, const block *ref_block,
|
||||
block *next_block, int with_xor) {
|
||||
__m128i block_XY[ARGON2_OWORDS_IN_BLOCK];
|
||||
unsigned int i;
|
||||
|
||||
if (with_xor) {
|
||||
for (i = 0; i < ARGON2_OWORDS_IN_BLOCK; i++) {
|
||||
state[i] = _mm_xor_si128(
|
||||
state[i], _mm_loadu_si128((const __m128i *)ref_block->v + i));
|
||||
block_XY[i] = _mm_xor_si128(
|
||||
state[i], _mm_loadu_si128((const __m128i *)next_block->v + i));
|
||||
}
|
||||
} else {
|
||||
for (i = 0; i < ARGON2_OWORDS_IN_BLOCK; i++) {
|
||||
block_XY[i] = state[i] = _mm_xor_si128(
|
||||
state[i], _mm_loadu_si128((const __m128i *)ref_block->v + i));
|
||||
}
|
||||
}
|
||||
|
||||
BLAKE2_ROUND( state[ 0], state[ 1], state[ 2], state[ 3],
|
||||
state[ 4], state[ 5], state[ 6], state[ 7] );
|
||||
BLAKE2_ROUND( state[ 8], state[ 9], state[10], state[11],
|
||||
state[12], state[13], state[14], state[15] );
|
||||
BLAKE2_ROUND( state[16], state[17], state[18], state[19],
|
||||
state[20], state[21], state[22], state[23] );
|
||||
BLAKE2_ROUND( state[24], state[25], state[26], state[27],
|
||||
state[28], state[29], state[30], state[31] );
|
||||
BLAKE2_ROUND( state[32], state[33], state[34], state[35],
|
||||
state[36], state[37], state[38], state[39] );
|
||||
BLAKE2_ROUND( state[40], state[41], state[42], state[43],
|
||||
state[44], state[45], state[46], state[47] );
|
||||
BLAKE2_ROUND( state[48], state[49], state[50], state[51],
|
||||
state[52], state[53], state[54], state[55] );
|
||||
BLAKE2_ROUND( state[56], state[57], state[58], state[59],
|
||||
state[60], state[61], state[62], state[63] );
|
||||
|
||||
BLAKE2_ROUND( state[ 0], state[ 8], state[16], state[24],
|
||||
state[32], state[40], state[48], state[56] );
|
||||
BLAKE2_ROUND( state[ 1], state[ 9], state[17], state[25],
|
||||
state[33], state[41], state[49], state[57] );
|
||||
BLAKE2_ROUND( state[ 2], state[10], state[18], state[26],
|
||||
state[34], state[42], state[50], state[58] );
|
||||
BLAKE2_ROUND( state[ 3], state[11], state[19], state[27],
|
||||
state[35], state[43], state[51], state[59] );
|
||||
BLAKE2_ROUND( state[ 4], state[12], state[20], state[28],
|
||||
state[36], state[44], state[52], state[60] );
|
||||
BLAKE2_ROUND( state[ 5], state[13], state[21], state[29],
|
||||
state[37], state[45], state[53], state[61] );
|
||||
BLAKE2_ROUND( state[ 6], state[14], state[22], state[30],
|
||||
state[38], state[46], state[54], state[62] );
|
||||
BLAKE2_ROUND( state[ 7], state[15], state[23], state[31],
|
||||
state[39], state[47], state[55], state[63] );
|
||||
|
||||
/*
|
||||
for (i = 0; i < 8; ++i) {
|
||||
BLAKE2_ROUND(state[8 * i + 0], state[8 * i + 1], state[8 * i + 2],
|
||||
state[8 * i + 3], state[8 * i + 4], state[8 * i + 5],
|
||||
state[8 * i + 6], state[8 * i + 7]);
|
||||
}
|
||||
|
||||
for (i = 0; i < 8; ++i) {
|
||||
BLAKE2_ROUND(state[8 * 0 + i], state[8 * 1 + i], state[8 * 2 + i],
|
||||
state[8 * 3 + i], state[8 * 4 + i], state[8 * 5 + i],
|
||||
state[8 * 6 + i], state[8 * 7 + i]);
|
||||
}
|
||||
*/
|
||||
for (i = 0; i < ARGON2_OWORDS_IN_BLOCK; i++) {
|
||||
state[i] = _mm_xor_si128(state[i], block_XY[i]);
|
||||
_mm_storeu_si128((__m128i *)next_block->v + i, state[i]);
|
||||
}
|
||||
}
|
||||
|
||||
#endif
|
||||
|
||||
#if 0
|
||||
static void next_addresses(block *address_block, block *input_block) {
|
||||
/*Temporary zero-initialized blocks*/
|
||||
#if defined(__AVX512F__)
|
||||
__m512i zero_block[ARGON2_512BIT_WORDS_IN_BLOCK];
|
||||
__m512i zero2_block[ARGON2_512BIT_WORDS_IN_BLOCK];
|
||||
#elif defined(__AVX2__)
|
||||
__m256i zero_block[ARGON2_HWORDS_IN_BLOCK];
|
||||
__m256i zero2_block[ARGON2_HWORDS_IN_BLOCK];
|
||||
#else
|
||||
__m128i zero_block[ARGON2_OWORDS_IN_BLOCK];
|
||||
__m128i zero2_block[ARGON2_OWORDS_IN_BLOCK];
|
||||
#endif
|
||||
|
||||
memset(zero_block, 0, sizeof(zero_block));
|
||||
memset(zero2_block, 0, sizeof(zero2_block));
|
||||
|
||||
/*Increasing index counter*/
|
||||
input_block->v[6]++;
|
||||
|
||||
/*First iteration of G*/
|
||||
fill_block(zero_block, input_block, address_block, 0);
|
||||
|
||||
/*Second iteration of G*/
|
||||
fill_block(zero2_block, address_block, address_block, 0);
|
||||
}
|
||||
#endif
|
||||
|
||||
void fill_segment(const argon2_instance_t *instance,
|
||||
argon2_position_t position) {
|
||||
block *ref_block = NULL, *curr_block = NULL;
|
||||
// block address_block, input_block;
|
||||
uint64_t pseudo_rand, ref_index, ref_lane;
|
||||
uint32_t prev_offset, curr_offset;
|
||||
uint32_t starting_index, i;
|
||||
#if defined(__AVX512F__)
|
||||
__m512i state[ARGON2_512BIT_WORDS_IN_BLOCK];
|
||||
#elif defined(__AVX2__)
|
||||
__m256i state[ARGON2_HWORDS_IN_BLOCK];
|
||||
#else
|
||||
__m128i state[ARGON2_OWORDS_IN_BLOCK];
|
||||
#endif
|
||||
// int data_independent_addressing;
|
||||
|
||||
if (instance == NULL) {
|
||||
return;
|
||||
}
|
||||
|
||||
// data_independent_addressing =
|
||||
// (instance->type == Argon2_i) ||
|
||||
// (instance->type == Argon2_id && (position.pass == 0) &&
|
||||
// (position.slice < ARGON2_SYNC_POINTS / 2));
|
||||
|
||||
// if (data_independent_addressing) {
|
||||
// init_block_value(&input_block, 0);
|
||||
|
||||
// input_block.v[0] = position.pass;
|
||||
// input_block.v[1] = position.lane;
|
||||
// input_block.v[2] = position.slice;
|
||||
// input_block.v[3] = instance->memory_blocks;
|
||||
// input_block.v[4] = instance->passes;
|
||||
// input_block.v[5] = instance->type;
|
||||
// }
|
||||
|
||||
starting_index = 0;
|
||||
|
||||
if ((0 == position.pass) && (0 == position.slice)) {
|
||||
starting_index = 2; /* we have already generated the first two blocks */
|
||||
|
||||
/* Don't forget to generate the first block of addresses: */
|
||||
// if (data_independent_addressing) {
|
||||
// next_addresses(&address_block, &input_block);
|
||||
// }
|
||||
}
|
||||
|
||||
/* Offset of the current block */
|
||||
curr_offset = position.lane * instance->lane_length +
|
||||
position.slice * instance->segment_length + starting_index;
|
||||
|
||||
if (0 == curr_offset % instance->lane_length) {
|
||||
/* Last block in this lane */
|
||||
prev_offset = curr_offset + instance->lane_length - 1;
|
||||
} else {
|
||||
/* Previous block */
|
||||
prev_offset = curr_offset - 1;
|
||||
}
|
||||
|
||||
memcpy(state, ((instance->memory + prev_offset)->v), ARGON2_BLOCK_SIZE);
|
||||
|
||||
for (i = starting_index; i < instance->segment_length;
|
||||
++i, ++curr_offset, ++prev_offset) {
|
||||
/*1.1 Rotating prev_offset if needed */
|
||||
if (curr_offset % instance->lane_length == 1) {
|
||||
prev_offset = curr_offset - 1;
|
||||
}
|
||||
|
||||
/* 1.2 Computing the index of the reference block */
|
||||
/* 1.2.1 Taking pseudo-random value from the previous block */
|
||||
// if (data_independent_addressing) {
|
||||
// if (i % ARGON2_ADDRESSES_IN_BLOCK == 0) {
|
||||
// next_addresses(&address_block, &input_block);
|
||||
// }
|
||||
// pseudo_rand = address_block.v[i % ARGON2_ADDRESSES_IN_BLOCK];
|
||||
// } else {
|
||||
pseudo_rand = instance->memory[prev_offset].v[0];
|
||||
// }
|
||||
|
||||
/* 1.2.2 Computing the lane of the reference block */
|
||||
ref_lane = ((pseudo_rand >> 32)) % instance->lanes;
|
||||
|
||||
if ((position.pass == 0) && (position.slice == 0)) {
|
||||
/* Can not reference other lanes yet */
|
||||
ref_lane = position.lane;
|
||||
}
|
||||
|
||||
/* 1.2.3 Computing the number of possible reference block within the
|
||||
* lane.
|
||||
*/
|
||||
position.index = i;
|
||||
ref_index = index_alpha(instance, &position, pseudo_rand & 0xFFFFFFFF,
|
||||
ref_lane == position.lane);
|
||||
|
||||
/* 2 Creating a new block */
|
||||
ref_block =
|
||||
instance->memory + instance->lane_length * ref_lane + ref_index;
|
||||
curr_block = instance->memory + curr_offset;
|
||||
if (ARGON2_VERSION_10 == instance->version) {
|
||||
/* version 1.2.1 and earlier: overwrite, not XOR */
|
||||
fill_block(state, ref_block, curr_block, 0);
|
||||
} else {
|
||||
if(0 == position.pass) {
|
||||
fill_block(state, ref_block, curr_block, 0);
|
||||
} else {
|
||||
fill_block(state, ref_block, curr_block, 1);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
156
algo/argon2/argon2d/blake2/blake2-impl.h
Normal file
156
algo/argon2/argon2d/blake2/blake2-impl.h
Normal file
@@ -0,0 +1,156 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#ifndef PORTABLE_BLAKE2_IMPL_H
|
||||
#define PORTABLE_BLAKE2_IMPL_H
|
||||
|
||||
#include <stdint.h>
|
||||
#include <string.h>
|
||||
|
||||
#if defined(_MSC_VER)
|
||||
#define BLAKE2_INLINE __inline
|
||||
#elif defined(__GNUC__) || defined(__clang__)
|
||||
#define BLAKE2_INLINE __inline__
|
||||
#else
|
||||
#define BLAKE2_INLINE
|
||||
#endif
|
||||
|
||||
/* Argon2 Team - Begin Code */
|
||||
/*
|
||||
Not an exhaustive list, but should cover the majority of modern platforms
|
||||
Additionally, the code will always be correct---this is only a performance
|
||||
tweak.
|
||||
*/
|
||||
#if (defined(__BYTE_ORDER__) && \
|
||||
(__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)) || \
|
||||
defined(__LITTLE_ENDIAN__) || defined(__ARMEL__) || defined(__MIPSEL__) || \
|
||||
defined(__AARCH64EL__) || defined(__amd64__) || defined(__i386__) || \
|
||||
defined(_M_IX86) || defined(_M_X64) || defined(_M_AMD64) || \
|
||||
defined(_M_ARM)
|
||||
#define NATIVE_LITTLE_ENDIAN
|
||||
#endif
|
||||
/* Argon2 Team - End Code */
|
||||
|
||||
static BLAKE2_INLINE uint32_t load32(const void *src) {
|
||||
#if defined(NATIVE_LITTLE_ENDIAN)
|
||||
uint32_t w;
|
||||
memcpy(&w, src, sizeof w);
|
||||
return w;
|
||||
#else
|
||||
const uint8_t *p = (const uint8_t *)src;
|
||||
uint32_t w = *p++;
|
||||
w |= (uint32_t)(*p++) << 8;
|
||||
w |= (uint32_t)(*p++) << 16;
|
||||
w |= (uint32_t)(*p++) << 24;
|
||||
return w;
|
||||
#endif
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE uint64_t load64(const void *src) {
|
||||
#if defined(NATIVE_LITTLE_ENDIAN)
|
||||
uint64_t w;
|
||||
memcpy(&w, src, sizeof w);
|
||||
return w;
|
||||
#else
|
||||
const uint8_t *p = (const uint8_t *)src;
|
||||
uint64_t w = *p++;
|
||||
w |= (uint64_t)(*p++) << 8;
|
||||
w |= (uint64_t)(*p++) << 16;
|
||||
w |= (uint64_t)(*p++) << 24;
|
||||
w |= (uint64_t)(*p++) << 32;
|
||||
w |= (uint64_t)(*p++) << 40;
|
||||
w |= (uint64_t)(*p++) << 48;
|
||||
w |= (uint64_t)(*p++) << 56;
|
||||
return w;
|
||||
#endif
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE void store32(void *dst, uint32_t w) {
|
||||
#if defined(NATIVE_LITTLE_ENDIAN)
|
||||
memcpy(dst, &w, sizeof w);
|
||||
#else
|
||||
uint8_t *p = (uint8_t *)dst;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
#endif
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE void store64(void *dst, uint64_t w) {
|
||||
#if defined(NATIVE_LITTLE_ENDIAN)
|
||||
memcpy(dst, &w, sizeof w);
|
||||
#else
|
||||
uint8_t *p = (uint8_t *)dst;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
#endif
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE uint64_t load48(const void *src) {
|
||||
const uint8_t *p = (const uint8_t *)src;
|
||||
uint64_t w = *p++;
|
||||
w |= (uint64_t)(*p++) << 8;
|
||||
w |= (uint64_t)(*p++) << 16;
|
||||
w |= (uint64_t)(*p++) << 24;
|
||||
w |= (uint64_t)(*p++) << 32;
|
||||
w |= (uint64_t)(*p++) << 40;
|
||||
return w;
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE void store48(void *dst, uint64_t w) {
|
||||
uint8_t *p = (uint8_t *)dst;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
w >>= 8;
|
||||
*p++ = (uint8_t)w;
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE uint32_t rotr32(const uint32_t w, const unsigned c) {
|
||||
return (w >> c) | (w << (32 - c));
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE uint64_t rotr64(const uint64_t w, const unsigned c) {
|
||||
return (w >> c) | (w << (64 - c));
|
||||
}
|
||||
|
||||
void clear_internal_memory(void *v, size_t n);
|
||||
|
||||
#endif
|
||||
91
algo/argon2/argon2d/blake2/blake2.h
Normal file
91
algo/argon2/argon2d/blake2/blake2.h
Normal file
@@ -0,0 +1,91 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#ifndef PORTABLE_BLAKE2_H
|
||||
#define PORTABLE_BLAKE2_H
|
||||
|
||||
#include <stddef.h>
|
||||
#include <stdint.h>
|
||||
#include <limits.h>
|
||||
|
||||
#if defined(__cplusplus)
|
||||
extern "C" {
|
||||
#endif
|
||||
|
||||
enum blake2b_constant {
|
||||
BLAKE2B_BLOCKBYTES = 128,
|
||||
BLAKE2B_OUTBYTES = 64,
|
||||
BLAKE2B_KEYBYTES = 64,
|
||||
BLAKE2B_SALTBYTES = 16,
|
||||
BLAKE2B_PERSONALBYTES = 16
|
||||
};
|
||||
|
||||
#pragma pack(push, 1)
|
||||
typedef struct __blake2b_param {
|
||||
uint8_t digest_length; /* 1 */
|
||||
uint8_t key_length; /* 2 */
|
||||
uint8_t fanout; /* 3 */
|
||||
uint8_t depth; /* 4 */
|
||||
uint32_t leaf_length; /* 8 */
|
||||
uint64_t node_offset; /* 16 */
|
||||
uint8_t node_depth; /* 17 */
|
||||
uint8_t inner_length; /* 18 */
|
||||
uint8_t reserved[14]; /* 32 */
|
||||
uint8_t salt[BLAKE2B_SALTBYTES]; /* 48 */
|
||||
uint8_t personal[BLAKE2B_PERSONALBYTES]; /* 64 */
|
||||
} blake2b_param;
|
||||
#pragma pack(pop)
|
||||
|
||||
typedef struct __blake2b_state {
|
||||
uint64_t h[8];
|
||||
uint64_t t[2];
|
||||
uint64_t f[2];
|
||||
uint8_t buf[BLAKE2B_BLOCKBYTES];
|
||||
unsigned buflen;
|
||||
unsigned outlen;
|
||||
uint8_t last_node;
|
||||
} blake2b_state;
|
||||
|
||||
/* Ensure param structs have not been wrongly padded */
|
||||
/* Poor man's static_assert */
|
||||
enum {
|
||||
blake2_size_check_0 = 1 / !!(CHAR_BIT == 8),
|
||||
blake2_size_check_2 =
|
||||
1 / !!(sizeof(blake2b_param) == sizeof(uint64_t) * CHAR_BIT)
|
||||
};
|
||||
|
||||
/* Streaming API */
|
||||
int blake2b_init(blake2b_state *S, size_t outlen);
|
||||
int blake2b_init_key(blake2b_state *S, size_t outlen, const void *key,
|
||||
size_t keylen);
|
||||
int blake2b_init_param(blake2b_state *S, const blake2b_param *P);
|
||||
int blake2b_update(blake2b_state *S, const void *in, size_t inlen);
|
||||
int blake2b_final(blake2b_state *S, void *out, size_t outlen);
|
||||
|
||||
/* Simple API */
|
||||
int blake2b(void *out, size_t outlen, const void *in, size_t inlen,
|
||||
const void *key, size_t keylen);
|
||||
|
||||
/* Argon2 Team - Begin Code */
|
||||
int blake2b_long(void *out, size_t outlen, const void *in, size_t inlen);
|
||||
/* Argon2 Team - End Code */
|
||||
|
||||
#if defined(__cplusplus)
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif
|
||||
390
algo/argon2/argon2d/blake2/blake2b.c
Normal file
390
algo/argon2/argon2d/blake2/blake2b.c
Normal file
@@ -0,0 +1,390 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#include <stdint.h>
|
||||
#include <string.h>
|
||||
#include <stdio.h>
|
||||
|
||||
#include "blake2.h"
|
||||
#include "blake2-impl.h"
|
||||
|
||||
static const uint64_t blake2b_IV[8] = {
|
||||
UINT64_C(0x6a09e667f3bcc908), UINT64_C(0xbb67ae8584caa73b),
|
||||
UINT64_C(0x3c6ef372fe94f82b), UINT64_C(0xa54ff53a5f1d36f1),
|
||||
UINT64_C(0x510e527fade682d1), UINT64_C(0x9b05688c2b3e6c1f),
|
||||
UINT64_C(0x1f83d9abfb41bd6b), UINT64_C(0x5be0cd19137e2179)};
|
||||
|
||||
static const unsigned int blake2b_sigma[12][16] = {
|
||||
{0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15},
|
||||
{14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3},
|
||||
{11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4},
|
||||
{7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8},
|
||||
{9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13},
|
||||
{2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9},
|
||||
{12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11},
|
||||
{13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10},
|
||||
{6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5},
|
||||
{10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13, 0},
|
||||
{0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15},
|
||||
{14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3},
|
||||
};
|
||||
|
||||
static BLAKE2_INLINE void blake2b_set_lastnode(blake2b_state *S) {
|
||||
S->f[1] = (uint64_t)-1;
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE void blake2b_set_lastblock(blake2b_state *S) {
|
||||
if (S->last_node) {
|
||||
blake2b_set_lastnode(S);
|
||||
}
|
||||
S->f[0] = (uint64_t)-1;
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE void blake2b_increment_counter(blake2b_state *S,
|
||||
uint64_t inc) {
|
||||
S->t[0] += inc;
|
||||
S->t[1] += (S->t[0] < inc);
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE void blake2b_invalidate_state(blake2b_state *S) {
|
||||
clear_internal_memory(S, sizeof(*S)); /* wipe */
|
||||
blake2b_set_lastblock(S); /* invalidate for further use */
|
||||
}
|
||||
|
||||
static BLAKE2_INLINE void blake2b_init0(blake2b_state *S) {
|
||||
memset(S, 0, sizeof(*S));
|
||||
memcpy(S->h, blake2b_IV, sizeof(S->h));
|
||||
}
|
||||
|
||||
int blake2b_init_param(blake2b_state *S, const blake2b_param *P) {
|
||||
const unsigned char *p = (const unsigned char *)P;
|
||||
unsigned int i;
|
||||
|
||||
if (NULL == P || NULL == S) {
|
||||
return -1;
|
||||
}
|
||||
|
||||
blake2b_init0(S);
|
||||
/* IV XOR Parameter Block */
|
||||
for (i = 0; i < 8; ++i) {
|
||||
S->h[i] ^= load64(&p[i * sizeof(S->h[i])]);
|
||||
}
|
||||
S->outlen = P->digest_length;
|
||||
return 0;
|
||||
}
|
||||
|
||||
/* Sequential blake2b initialization */
|
||||
int blake2b_init(blake2b_state *S, size_t outlen) {
|
||||
blake2b_param P;
|
||||
|
||||
if (S == NULL) {
|
||||
return -1;
|
||||
}
|
||||
|
||||
if ((outlen == 0) || (outlen > BLAKE2B_OUTBYTES)) {
|
||||
blake2b_invalidate_state(S);
|
||||
return -1;
|
||||
}
|
||||
|
||||
/* Setup Parameter Block for unkeyed BLAKE2 */
|
||||
P.digest_length = (uint8_t)outlen;
|
||||
P.key_length = 0;
|
||||
P.fanout = 1;
|
||||
P.depth = 1;
|
||||
P.leaf_length = 0;
|
||||
P.node_offset = 0;
|
||||
P.node_depth = 0;
|
||||
P.inner_length = 0;
|
||||
memset(P.reserved, 0, sizeof(P.reserved));
|
||||
memset(P.salt, 0, sizeof(P.salt));
|
||||
memset(P.personal, 0, sizeof(P.personal));
|
||||
|
||||
return blake2b_init_param(S, &P);
|
||||
}
|
||||
|
||||
int blake2b_init_key(blake2b_state *S, size_t outlen, const void *key,
|
||||
size_t keylen) {
|
||||
blake2b_param P;
|
||||
|
||||
if (S == NULL) {
|
||||
return -1;
|
||||
}
|
||||
|
||||
if ((outlen == 0) || (outlen > BLAKE2B_OUTBYTES)) {
|
||||
blake2b_invalidate_state(S);
|
||||
return -1;
|
||||
}
|
||||
|
||||
if ((key == 0) || (keylen == 0) || (keylen > BLAKE2B_KEYBYTES)) {
|
||||
blake2b_invalidate_state(S);
|
||||
return -1;
|
||||
}
|
||||
|
||||
/* Setup Parameter Block for keyed BLAKE2 */
|
||||
P.digest_length = (uint8_t)outlen;
|
||||
P.key_length = (uint8_t)keylen;
|
||||
P.fanout = 1;
|
||||
P.depth = 1;
|
||||
P.leaf_length = 0;
|
||||
P.node_offset = 0;
|
||||
P.node_depth = 0;
|
||||
P.inner_length = 0;
|
||||
memset(P.reserved, 0, sizeof(P.reserved));
|
||||
memset(P.salt, 0, sizeof(P.salt));
|
||||
memset(P.personal, 0, sizeof(P.personal));
|
||||
|
||||
if (blake2b_init_param(S, &P) < 0) {
|
||||
blake2b_invalidate_state(S);
|
||||
return -1;
|
||||
}
|
||||
|
||||
{
|
||||
uint8_t block[BLAKE2B_BLOCKBYTES];
|
||||
memset(block, 0, BLAKE2B_BLOCKBYTES);
|
||||
memcpy(block, key, keylen);
|
||||
blake2b_update(S, block, BLAKE2B_BLOCKBYTES);
|
||||
/* Burn the key from stack */
|
||||
clear_internal_memory(block, BLAKE2B_BLOCKBYTES);
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
|
||||
static void blake2b_compress(blake2b_state *S, const uint8_t *block) {
|
||||
uint64_t m[16];
|
||||
uint64_t v[16];
|
||||
unsigned int i, r;
|
||||
|
||||
for (i = 0; i < 16; ++i) {
|
||||
m[i] = load64(block + i * sizeof(m[i]));
|
||||
}
|
||||
|
||||
for (i = 0; i < 8; ++i) {
|
||||
v[i] = S->h[i];
|
||||
}
|
||||
|
||||
v[8] = blake2b_IV[0];
|
||||
v[9] = blake2b_IV[1];
|
||||
v[10] = blake2b_IV[2];
|
||||
v[11] = blake2b_IV[3];
|
||||
v[12] = blake2b_IV[4] ^ S->t[0];
|
||||
v[13] = blake2b_IV[5] ^ S->t[1];
|
||||
v[14] = blake2b_IV[6] ^ S->f[0];
|
||||
v[15] = blake2b_IV[7] ^ S->f[1];
|
||||
|
||||
#define G(r, i, a, b, c, d) \
|
||||
do { \
|
||||
a = a + b + m[blake2b_sigma[r][2 * i + 0]]; \
|
||||
d = rotr64(d ^ a, 32); \
|
||||
c = c + d; \
|
||||
b = rotr64(b ^ c, 24); \
|
||||
a = a + b + m[blake2b_sigma[r][2 * i + 1]]; \
|
||||
d = rotr64(d ^ a, 16); \
|
||||
c = c + d; \
|
||||
b = rotr64(b ^ c, 63); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define ROUND(r) \
|
||||
do { \
|
||||
G(r, 0, v[0], v[4], v[8], v[12]); \
|
||||
G(r, 1, v[1], v[5], v[9], v[13]); \
|
||||
G(r, 2, v[2], v[6], v[10], v[14]); \
|
||||
G(r, 3, v[3], v[7], v[11], v[15]); \
|
||||
G(r, 4, v[0], v[5], v[10], v[15]); \
|
||||
G(r, 5, v[1], v[6], v[11], v[12]); \
|
||||
G(r, 6, v[2], v[7], v[8], v[13]); \
|
||||
G(r, 7, v[3], v[4], v[9], v[14]); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
for (r = 0; r < 12; ++r) {
|
||||
ROUND(r);
|
||||
}
|
||||
|
||||
for (i = 0; i < 8; ++i) {
|
||||
S->h[i] = S->h[i] ^ v[i] ^ v[i + 8];
|
||||
}
|
||||
|
||||
#undef G
|
||||
#undef ROUND
|
||||
}
|
||||
|
||||
int blake2b_update(blake2b_state *S, const void *in, size_t inlen) {
|
||||
const uint8_t *pin = (const uint8_t *)in;
|
||||
|
||||
if (inlen == 0) {
|
||||
return 0;
|
||||
}
|
||||
|
||||
/* Sanity check */
|
||||
if (S == NULL || in == NULL) {
|
||||
return -1;
|
||||
}
|
||||
|
||||
/* Is this a reused state? */
|
||||
if (S->f[0] != 0) {
|
||||
return -1;
|
||||
}
|
||||
|
||||
if (S->buflen + inlen > BLAKE2B_BLOCKBYTES) {
|
||||
/* Complete current block */
|
||||
size_t left = S->buflen;
|
||||
size_t fill = BLAKE2B_BLOCKBYTES - left;
|
||||
memcpy(&S->buf[left], pin, fill);
|
||||
blake2b_increment_counter(S, BLAKE2B_BLOCKBYTES);
|
||||
blake2b_compress(S, S->buf);
|
||||
S->buflen = 0;
|
||||
inlen -= fill;
|
||||
pin += fill;
|
||||
/* Avoid buffer copies when possible */
|
||||
while (inlen > BLAKE2B_BLOCKBYTES) {
|
||||
blake2b_increment_counter(S, BLAKE2B_BLOCKBYTES);
|
||||
blake2b_compress(S, pin);
|
||||
inlen -= BLAKE2B_BLOCKBYTES;
|
||||
pin += BLAKE2B_BLOCKBYTES;
|
||||
}
|
||||
}
|
||||
memcpy(&S->buf[S->buflen], pin, inlen);
|
||||
S->buflen += (unsigned int)inlen;
|
||||
return 0;
|
||||
}
|
||||
|
||||
int blake2b_final(blake2b_state *S, void *out, size_t outlen) {
|
||||
uint8_t buffer[BLAKE2B_OUTBYTES] = {0};
|
||||
unsigned int i;
|
||||
|
||||
/* Sanity checks */
|
||||
if (S == NULL || out == NULL || outlen < S->outlen) {
|
||||
return -1;
|
||||
}
|
||||
|
||||
/* Is this a reused state? */
|
||||
if (S->f[0] != 0) {
|
||||
return -1;
|
||||
}
|
||||
|
||||
blake2b_increment_counter(S, S->buflen);
|
||||
blake2b_set_lastblock(S);
|
||||
memset(&S->buf[S->buflen], 0, BLAKE2B_BLOCKBYTES - S->buflen); /* Padding */
|
||||
blake2b_compress(S, S->buf);
|
||||
|
||||
for (i = 0; i < 8; ++i) { /* Output full hash to temp buffer */
|
||||
store64(buffer + sizeof(S->h[i]) * i, S->h[i]);
|
||||
}
|
||||
|
||||
memcpy(out, buffer, S->outlen);
|
||||
clear_internal_memory(buffer, sizeof(buffer));
|
||||
clear_internal_memory(S->buf, sizeof(S->buf));
|
||||
clear_internal_memory(S->h, sizeof(S->h));
|
||||
return 0;
|
||||
}
|
||||
|
||||
int blake2b(void *out, size_t outlen, const void *in, size_t inlen,
|
||||
const void *key, size_t keylen) {
|
||||
blake2b_state S;
|
||||
int ret = -1;
|
||||
|
||||
/* Verify parameters */
|
||||
if (NULL == in && inlen > 0) {
|
||||
goto fail;
|
||||
}
|
||||
|
||||
if (NULL == out || outlen == 0 || outlen > BLAKE2B_OUTBYTES) {
|
||||
goto fail;
|
||||
}
|
||||
|
||||
if ((NULL == key && keylen > 0) || keylen > BLAKE2B_KEYBYTES) {
|
||||
goto fail;
|
||||
}
|
||||
|
||||
if (keylen > 0) {
|
||||
if (blake2b_init_key(&S, outlen, key, keylen) < 0) {
|
||||
goto fail;
|
||||
}
|
||||
} else {
|
||||
if (blake2b_init(&S, outlen) < 0) {
|
||||
goto fail;
|
||||
}
|
||||
}
|
||||
|
||||
if (blake2b_update(&S, in, inlen) < 0) {
|
||||
goto fail;
|
||||
}
|
||||
ret = blake2b_final(&S, out, outlen);
|
||||
|
||||
fail:
|
||||
clear_internal_memory(&S, sizeof(S));
|
||||
return ret;
|
||||
}
|
||||
|
||||
/* Argon2 Team - Begin Code */
|
||||
int blake2b_long(void *pout, size_t outlen, const void *in, size_t inlen) {
|
||||
uint8_t *out = (uint8_t *)pout;
|
||||
blake2b_state blake_state;
|
||||
uint8_t outlen_bytes[sizeof(uint32_t)] = {0};
|
||||
int ret = -1;
|
||||
|
||||
if (outlen > UINT32_MAX) {
|
||||
goto fail;
|
||||
}
|
||||
|
||||
/* Ensure little-endian byte order! */
|
||||
store32(outlen_bytes, (uint32_t)outlen);
|
||||
|
||||
#define TRY(statement) \
|
||||
do { \
|
||||
ret = statement; \
|
||||
if (ret < 0) { \
|
||||
goto fail; \
|
||||
} \
|
||||
} while ((void)0, 0)
|
||||
|
||||
if (outlen <= BLAKE2B_OUTBYTES) {
|
||||
TRY(blake2b_init(&blake_state, outlen));
|
||||
TRY(blake2b_update(&blake_state, outlen_bytes, sizeof(outlen_bytes)));
|
||||
TRY(blake2b_update(&blake_state, in, inlen));
|
||||
TRY(blake2b_final(&blake_state, out, outlen));
|
||||
} else {
|
||||
uint32_t toproduce;
|
||||
uint8_t out_buffer[BLAKE2B_OUTBYTES];
|
||||
uint8_t in_buffer[BLAKE2B_OUTBYTES];
|
||||
TRY(blake2b_init(&blake_state, BLAKE2B_OUTBYTES));
|
||||
TRY(blake2b_update(&blake_state, outlen_bytes, sizeof(outlen_bytes)));
|
||||
TRY(blake2b_update(&blake_state, in, inlen));
|
||||
TRY(blake2b_final(&blake_state, out_buffer, BLAKE2B_OUTBYTES));
|
||||
memcpy(out, out_buffer, BLAKE2B_OUTBYTES / 2);
|
||||
out += BLAKE2B_OUTBYTES / 2;
|
||||
toproduce = (uint32_t)outlen - BLAKE2B_OUTBYTES / 2;
|
||||
|
||||
while (toproduce > BLAKE2B_OUTBYTES) {
|
||||
memcpy(in_buffer, out_buffer, BLAKE2B_OUTBYTES);
|
||||
TRY(blake2b(out_buffer, BLAKE2B_OUTBYTES, in_buffer,
|
||||
BLAKE2B_OUTBYTES, NULL, 0));
|
||||
memcpy(out, out_buffer, BLAKE2B_OUTBYTES / 2);
|
||||
out += BLAKE2B_OUTBYTES / 2;
|
||||
toproduce -= BLAKE2B_OUTBYTES / 2;
|
||||
}
|
||||
|
||||
memcpy(in_buffer, out_buffer, BLAKE2B_OUTBYTES);
|
||||
TRY(blake2b(out_buffer, toproduce, in_buffer, BLAKE2B_OUTBYTES, NULL,
|
||||
0));
|
||||
memcpy(out, out_buffer, toproduce);
|
||||
}
|
||||
fail:
|
||||
clear_internal_memory(&blake_state, sizeof(blake_state));
|
||||
return ret;
|
||||
#undef TRY
|
||||
}
|
||||
/* Argon2 Team - End Code */
|
||||
471
algo/argon2/argon2d/blake2/blamka-round-opt.h
Normal file
471
algo/argon2/argon2d/blake2/blamka-round-opt.h
Normal file
@@ -0,0 +1,471 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#ifndef BLAKE_ROUND_MKA_OPT_H
|
||||
#define BLAKE_ROUND_MKA_OPT_H
|
||||
|
||||
#include "blake2-impl.h"
|
||||
|
||||
#include <emmintrin.h>
|
||||
#if defined(__SSSE3__)
|
||||
#include <tmmintrin.h> /* for _mm_shuffle_epi8 and _mm_alignr_epi8 */
|
||||
#endif
|
||||
|
||||
#if defined(__XOP__) && (defined(__GNUC__) || defined(__clang__))
|
||||
#include <x86intrin.h>
|
||||
#endif
|
||||
|
||||
#if !defined(__AVX512F__)
|
||||
#if !defined(__AVX2__)
|
||||
#if !defined(__XOP__)
|
||||
#if defined(__SSSE3__)
|
||||
#define r16 \
|
||||
(_mm_setr_epi8(2, 3, 4, 5, 6, 7, 0, 1, 10, 11, 12, 13, 14, 15, 8, 9))
|
||||
#define r24 \
|
||||
(_mm_setr_epi8(3, 4, 5, 6, 7, 0, 1, 2, 11, 12, 13, 14, 15, 8, 9, 10))
|
||||
#define _mm_roti_epi64(x, c) \
|
||||
(-(c) == 32) \
|
||||
? _mm_shuffle_epi32((x), _MM_SHUFFLE(2, 3, 0, 1)) \
|
||||
: (-(c) == 24) \
|
||||
? _mm_shuffle_epi8((x), r24) \
|
||||
: (-(c) == 16) \
|
||||
? _mm_shuffle_epi8((x), r16) \
|
||||
: (-(c) == 63) \
|
||||
? _mm_xor_si128(_mm_srli_epi64((x), -(c)), \
|
||||
_mm_add_epi64((x), (x))) \
|
||||
: _mm_xor_si128(_mm_srli_epi64((x), -(c)), \
|
||||
_mm_slli_epi64((x), 64 - (-(c))))
|
||||
#else /* defined(__SSE2__) */
|
||||
#define _mm_roti_epi64(r, c) \
|
||||
_mm_xor_si128(_mm_srli_epi64((r), -(c)), _mm_slli_epi64((r), 64 - (-(c))))
|
||||
#endif
|
||||
#else
|
||||
#endif
|
||||
|
||||
static BLAKE2_INLINE __m128i fBlaMka(__m128i x, __m128i y) {
|
||||
const __m128i z = _mm_mul_epu32(x, y);
|
||||
return _mm_add_epi64(_mm_add_epi64(x, y), _mm_add_epi64(z, z));
|
||||
}
|
||||
|
||||
#define G1(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
A0 = fBlaMka(A0, B0); \
|
||||
A1 = fBlaMka(A1, B1); \
|
||||
\
|
||||
D0 = _mm_xor_si128(D0, A0); \
|
||||
D1 = _mm_xor_si128(D1, A1); \
|
||||
\
|
||||
D0 = _mm_roti_epi64(D0, -32); \
|
||||
D1 = _mm_roti_epi64(D1, -32); \
|
||||
\
|
||||
C0 = fBlaMka(C0, D0); \
|
||||
C1 = fBlaMka(C1, D1); \
|
||||
\
|
||||
B0 = _mm_xor_si128(B0, C0); \
|
||||
B1 = _mm_xor_si128(B1, C1); \
|
||||
\
|
||||
B0 = _mm_roti_epi64(B0, -24); \
|
||||
B1 = _mm_roti_epi64(B1, -24); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define G2(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
A0 = fBlaMka(A0, B0); \
|
||||
A1 = fBlaMka(A1, B1); \
|
||||
\
|
||||
D0 = _mm_xor_si128(D0, A0); \
|
||||
D1 = _mm_xor_si128(D1, A1); \
|
||||
\
|
||||
D0 = _mm_roti_epi64(D0, -16); \
|
||||
D1 = _mm_roti_epi64(D1, -16); \
|
||||
\
|
||||
C0 = fBlaMka(C0, D0); \
|
||||
C1 = fBlaMka(C1, D1); \
|
||||
\
|
||||
B0 = _mm_xor_si128(B0, C0); \
|
||||
B1 = _mm_xor_si128(B1, C1); \
|
||||
\
|
||||
B0 = _mm_roti_epi64(B0, -63); \
|
||||
B1 = _mm_roti_epi64(B1, -63); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#if defined(__SSSE3__)
|
||||
#define DIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
__m128i t0 = _mm_alignr_epi8(B1, B0, 8); \
|
||||
__m128i t1 = _mm_alignr_epi8(B0, B1, 8); \
|
||||
B0 = t0; \
|
||||
B1 = t1; \
|
||||
\
|
||||
t0 = C0; \
|
||||
C0 = C1; \
|
||||
C1 = t0; \
|
||||
\
|
||||
t0 = _mm_alignr_epi8(D1, D0, 8); \
|
||||
t1 = _mm_alignr_epi8(D0, D1, 8); \
|
||||
D0 = t1; \
|
||||
D1 = t0; \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define UNDIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
__m128i t0 = _mm_alignr_epi8(B0, B1, 8); \
|
||||
__m128i t1 = _mm_alignr_epi8(B1, B0, 8); \
|
||||
B0 = t0; \
|
||||
B1 = t1; \
|
||||
\
|
||||
t0 = C0; \
|
||||
C0 = C1; \
|
||||
C1 = t0; \
|
||||
\
|
||||
t0 = _mm_alignr_epi8(D0, D1, 8); \
|
||||
t1 = _mm_alignr_epi8(D1, D0, 8); \
|
||||
D0 = t1; \
|
||||
D1 = t0; \
|
||||
} while ((void)0, 0)
|
||||
#else /* SSE2 */
|
||||
#define DIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
__m128i t0 = D0; \
|
||||
__m128i t1 = B0; \
|
||||
D0 = C0; \
|
||||
C0 = C1; \
|
||||
C1 = D0; \
|
||||
D0 = _mm_unpackhi_epi64(D1, _mm_unpacklo_epi64(t0, t0)); \
|
||||
D1 = _mm_unpackhi_epi64(t0, _mm_unpacklo_epi64(D1, D1)); \
|
||||
B0 = _mm_unpackhi_epi64(B0, _mm_unpacklo_epi64(B1, B1)); \
|
||||
B1 = _mm_unpackhi_epi64(B1, _mm_unpacklo_epi64(t1, t1)); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define UNDIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
__m128i t0, t1; \
|
||||
t0 = C0; \
|
||||
C0 = C1; \
|
||||
C1 = t0; \
|
||||
t0 = B0; \
|
||||
t1 = D0; \
|
||||
B0 = _mm_unpackhi_epi64(B1, _mm_unpacklo_epi64(B0, B0)); \
|
||||
B1 = _mm_unpackhi_epi64(t0, _mm_unpacklo_epi64(B1, B1)); \
|
||||
D0 = _mm_unpackhi_epi64(D0, _mm_unpacklo_epi64(D1, D1)); \
|
||||
D1 = _mm_unpackhi_epi64(D1, _mm_unpacklo_epi64(t1, t1)); \
|
||||
} while ((void)0, 0)
|
||||
#endif
|
||||
|
||||
#define BLAKE2_ROUND(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
do { \
|
||||
G1(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
G2(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
\
|
||||
DIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
\
|
||||
G1(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
G2(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
\
|
||||
UNDIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
} while ((void)0, 0)
|
||||
#else /* __AVX2__ */
|
||||
|
||||
#include <immintrin.h>
|
||||
|
||||
#define rotr32(x) _mm256_shuffle_epi32(x, _MM_SHUFFLE(2, 3, 0, 1))
|
||||
#define rotr24(x) _mm256_shuffle_epi8(x, _mm256_setr_epi8(3, 4, 5, 6, 7, 0, 1, 2, 11, 12, 13, 14, 15, 8, 9, 10, 3, 4, 5, 6, 7, 0, 1, 2, 11, 12, 13, 14, 15, 8, 9, 10))
|
||||
#define rotr16(x) _mm256_shuffle_epi8(x, _mm256_setr_epi8(2, 3, 4, 5, 6, 7, 0, 1, 10, 11, 12, 13, 14, 15, 8, 9, 2, 3, 4, 5, 6, 7, 0, 1, 10, 11, 12, 13, 14, 15, 8, 9))
|
||||
#define rotr63(x) _mm256_xor_si256(_mm256_srli_epi64((x), 63), _mm256_add_epi64((x), (x)))
|
||||
|
||||
#define G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
do { \
|
||||
__m256i ml = _mm256_mul_epu32(A0, B0); \
|
||||
ml = _mm256_add_epi64(ml, ml); \
|
||||
A0 = _mm256_add_epi64(A0, _mm256_add_epi64(B0, ml)); \
|
||||
D0 = _mm256_xor_si256(D0, A0); \
|
||||
D0 = rotr32(D0); \
|
||||
\
|
||||
ml = _mm256_mul_epu32(C0, D0); \
|
||||
ml = _mm256_add_epi64(ml, ml); \
|
||||
C0 = _mm256_add_epi64(C0, _mm256_add_epi64(D0, ml)); \
|
||||
\
|
||||
B0 = _mm256_xor_si256(B0, C0); \
|
||||
B0 = rotr24(B0); \
|
||||
\
|
||||
ml = _mm256_mul_epu32(A1, B1); \
|
||||
ml = _mm256_add_epi64(ml, ml); \
|
||||
A1 = _mm256_add_epi64(A1, _mm256_add_epi64(B1, ml)); \
|
||||
D1 = _mm256_xor_si256(D1, A1); \
|
||||
D1 = rotr32(D1); \
|
||||
\
|
||||
ml = _mm256_mul_epu32(C1, D1); \
|
||||
ml = _mm256_add_epi64(ml, ml); \
|
||||
C1 = _mm256_add_epi64(C1, _mm256_add_epi64(D1, ml)); \
|
||||
\
|
||||
B1 = _mm256_xor_si256(B1, C1); \
|
||||
B1 = rotr24(B1); \
|
||||
} while((void)0, 0);
|
||||
|
||||
#define G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
do { \
|
||||
__m256i ml = _mm256_mul_epu32(A0, B0); \
|
||||
ml = _mm256_add_epi64(ml, ml); \
|
||||
A0 = _mm256_add_epi64(A0, _mm256_add_epi64(B0, ml)); \
|
||||
D0 = _mm256_xor_si256(D0, A0); \
|
||||
D0 = rotr16(D0); \
|
||||
\
|
||||
ml = _mm256_mul_epu32(C0, D0); \
|
||||
ml = _mm256_add_epi64(ml, ml); \
|
||||
C0 = _mm256_add_epi64(C0, _mm256_add_epi64(D0, ml)); \
|
||||
B0 = _mm256_xor_si256(B0, C0); \
|
||||
B0 = rotr63(B0); \
|
||||
\
|
||||
ml = _mm256_mul_epu32(A1, B1); \
|
||||
ml = _mm256_add_epi64(ml, ml); \
|
||||
A1 = _mm256_add_epi64(A1, _mm256_add_epi64(B1, ml)); \
|
||||
D1 = _mm256_xor_si256(D1, A1); \
|
||||
D1 = rotr16(D1); \
|
||||
\
|
||||
ml = _mm256_mul_epu32(C1, D1); \
|
||||
ml = _mm256_add_epi64(ml, ml); \
|
||||
C1 = _mm256_add_epi64(C1, _mm256_add_epi64(D1, ml)); \
|
||||
B1 = _mm256_xor_si256(B1, C1); \
|
||||
B1 = rotr63(B1); \
|
||||
} while((void)0, 0);
|
||||
|
||||
#define DIAGONALIZE_1(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
B0 = _mm256_permute4x64_epi64(B0, _MM_SHUFFLE(0, 3, 2, 1)); \
|
||||
C0 = _mm256_permute4x64_epi64(C0, _MM_SHUFFLE(1, 0, 3, 2)); \
|
||||
D0 = _mm256_permute4x64_epi64(D0, _MM_SHUFFLE(2, 1, 0, 3)); \
|
||||
\
|
||||
B1 = _mm256_permute4x64_epi64(B1, _MM_SHUFFLE(0, 3, 2, 1)); \
|
||||
C1 = _mm256_permute4x64_epi64(C1, _MM_SHUFFLE(1, 0, 3, 2)); \
|
||||
D1 = _mm256_permute4x64_epi64(D1, _MM_SHUFFLE(2, 1, 0, 3)); \
|
||||
} while((void)0, 0);
|
||||
|
||||
#define DIAGONALIZE_2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
do { \
|
||||
__m256i tmp1 = _mm256_blend_epi32(B0, B1, 0xCC); \
|
||||
__m256i tmp2 = _mm256_blend_epi32(B0, B1, 0x33); \
|
||||
B1 = _mm256_permute4x64_epi64(tmp1, _MM_SHUFFLE(2,3,0,1)); \
|
||||
B0 = _mm256_permute4x64_epi64(tmp2, _MM_SHUFFLE(2,3,0,1)); \
|
||||
\
|
||||
tmp1 = C0; \
|
||||
C0 = C1; \
|
||||
C1 = tmp1; \
|
||||
\
|
||||
tmp1 = _mm256_blend_epi32(D0, D1, 0xCC); \
|
||||
tmp2 = _mm256_blend_epi32(D0, D1, 0x33); \
|
||||
D0 = _mm256_permute4x64_epi64(tmp1, _MM_SHUFFLE(2,3,0,1)); \
|
||||
D1 = _mm256_permute4x64_epi64(tmp2, _MM_SHUFFLE(2,3,0,1)); \
|
||||
} while(0);
|
||||
|
||||
#define UNDIAGONALIZE_1(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
B0 = _mm256_permute4x64_epi64(B0, _MM_SHUFFLE(2, 1, 0, 3)); \
|
||||
C0 = _mm256_permute4x64_epi64(C0, _MM_SHUFFLE(1, 0, 3, 2)); \
|
||||
D0 = _mm256_permute4x64_epi64(D0, _MM_SHUFFLE(0, 3, 2, 1)); \
|
||||
\
|
||||
B1 = _mm256_permute4x64_epi64(B1, _MM_SHUFFLE(2, 1, 0, 3)); \
|
||||
C1 = _mm256_permute4x64_epi64(C1, _MM_SHUFFLE(1, 0, 3, 2)); \
|
||||
D1 = _mm256_permute4x64_epi64(D1, _MM_SHUFFLE(0, 3, 2, 1)); \
|
||||
} while((void)0, 0);
|
||||
|
||||
#define UNDIAGONALIZE_2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
do { \
|
||||
__m256i tmp1 = _mm256_blend_epi32(B0, B1, 0xCC); \
|
||||
__m256i tmp2 = _mm256_blend_epi32(B0, B1, 0x33); \
|
||||
B0 = _mm256_permute4x64_epi64(tmp1, _MM_SHUFFLE(2,3,0,1)); \
|
||||
B1 = _mm256_permute4x64_epi64(tmp2, _MM_SHUFFLE(2,3,0,1)); \
|
||||
\
|
||||
tmp1 = C0; \
|
||||
C0 = C1; \
|
||||
C1 = tmp1; \
|
||||
\
|
||||
tmp1 = _mm256_blend_epi32(D0, D1, 0x33); \
|
||||
tmp2 = _mm256_blend_epi32(D0, D1, 0xCC); \
|
||||
D0 = _mm256_permute4x64_epi64(tmp1, _MM_SHUFFLE(2,3,0,1)); \
|
||||
D1 = _mm256_permute4x64_epi64(tmp2, _MM_SHUFFLE(2,3,0,1)); \
|
||||
} while((void)0, 0);
|
||||
|
||||
#define BLAKE2_ROUND_1(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
do{ \
|
||||
G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
\
|
||||
DIAGONALIZE_1(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
\
|
||||
G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
\
|
||||
UNDIAGONALIZE_1(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
} while((void)0, 0);
|
||||
|
||||
#define BLAKE2_ROUND_2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
do{ \
|
||||
G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
\
|
||||
DIAGONALIZE_2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
\
|
||||
G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
\
|
||||
UNDIAGONALIZE_2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
} while((void)0, 0);
|
||||
|
||||
#endif /* __AVX2__ */
|
||||
|
||||
#else /* __AVX512F__ */
|
||||
|
||||
#include <immintrin.h>
|
||||
|
||||
#define ror64(x, n) _mm512_ror_epi64((x), (n))
|
||||
|
||||
static __m512i muladd(__m512i x, __m512i y)
|
||||
{
|
||||
__m512i z = _mm512_mul_epu32(x, y);
|
||||
return _mm512_add_epi64(_mm512_add_epi64(x, y), _mm512_add_epi64(z, z));
|
||||
}
|
||||
|
||||
#define G1(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
A0 = muladd(A0, B0); \
|
||||
A1 = muladd(A1, B1); \
|
||||
\
|
||||
D0 = _mm512_xor_si512(D0, A0); \
|
||||
D1 = _mm512_xor_si512(D1, A1); \
|
||||
\
|
||||
D0 = ror64(D0, 32); \
|
||||
D1 = ror64(D1, 32); \
|
||||
\
|
||||
C0 = muladd(C0, D0); \
|
||||
C1 = muladd(C1, D1); \
|
||||
\
|
||||
B0 = _mm512_xor_si512(B0, C0); \
|
||||
B1 = _mm512_xor_si512(B1, C1); \
|
||||
\
|
||||
B0 = ror64(B0, 24); \
|
||||
B1 = ror64(B1, 24); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define G2(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
A0 = muladd(A0, B0); \
|
||||
A1 = muladd(A1, B1); \
|
||||
\
|
||||
D0 = _mm512_xor_si512(D0, A0); \
|
||||
D1 = _mm512_xor_si512(D1, A1); \
|
||||
\
|
||||
D0 = ror64(D0, 16); \
|
||||
D1 = ror64(D1, 16); \
|
||||
\
|
||||
C0 = muladd(C0, D0); \
|
||||
C1 = muladd(C1, D1); \
|
||||
\
|
||||
B0 = _mm512_xor_si512(B0, C0); \
|
||||
B1 = _mm512_xor_si512(B1, C1); \
|
||||
\
|
||||
B0 = ror64(B0, 63); \
|
||||
B1 = ror64(B1, 63); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define DIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
B0 = _mm512_permutex_epi64(B0, _MM_SHUFFLE(0, 3, 2, 1)); \
|
||||
B1 = _mm512_permutex_epi64(B1, _MM_SHUFFLE(0, 3, 2, 1)); \
|
||||
\
|
||||
C0 = _mm512_permutex_epi64(C0, _MM_SHUFFLE(1, 0, 3, 2)); \
|
||||
C1 = _mm512_permutex_epi64(C1, _MM_SHUFFLE(1, 0, 3, 2)); \
|
||||
\
|
||||
D0 = _mm512_permutex_epi64(D0, _MM_SHUFFLE(2, 1, 0, 3)); \
|
||||
D1 = _mm512_permutex_epi64(D1, _MM_SHUFFLE(2, 1, 0, 3)); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define UNDIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
B0 = _mm512_permutex_epi64(B0, _MM_SHUFFLE(2, 1, 0, 3)); \
|
||||
B1 = _mm512_permutex_epi64(B1, _MM_SHUFFLE(2, 1, 0, 3)); \
|
||||
\
|
||||
C0 = _mm512_permutex_epi64(C0, _MM_SHUFFLE(1, 0, 3, 2)); \
|
||||
C1 = _mm512_permutex_epi64(C1, _MM_SHUFFLE(1, 0, 3, 2)); \
|
||||
\
|
||||
D0 = _mm512_permutex_epi64(D0, _MM_SHUFFLE(0, 3, 2, 1)); \
|
||||
D1 = _mm512_permutex_epi64(D1, _MM_SHUFFLE(0, 3, 2, 1)); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define BLAKE2_ROUND(A0, B0, C0, D0, A1, B1, C1, D1) \
|
||||
do { \
|
||||
G1(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
G2(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
\
|
||||
DIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
\
|
||||
G1(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
G2(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
\
|
||||
UNDIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define SWAP_HALVES(A0, A1) \
|
||||
do { \
|
||||
__m512i t0, t1; \
|
||||
t0 = _mm512_shuffle_i64x2(A0, A1, _MM_SHUFFLE(1, 0, 1, 0)); \
|
||||
t1 = _mm512_shuffle_i64x2(A0, A1, _MM_SHUFFLE(3, 2, 3, 2)); \
|
||||
A0 = t0; \
|
||||
A1 = t1; \
|
||||
} while((void)0, 0)
|
||||
|
||||
#define SWAP_QUARTERS(A0, A1) \
|
||||
do { \
|
||||
SWAP_HALVES(A0, A1); \
|
||||
A0 = _mm512_permutexvar_epi64(_mm512_setr_epi64(0, 1, 4, 5, 2, 3, 6, 7), A0); \
|
||||
A1 = _mm512_permutexvar_epi64(_mm512_setr_epi64(0, 1, 4, 5, 2, 3, 6, 7), A1); \
|
||||
} while((void)0, 0)
|
||||
|
||||
#define UNSWAP_QUARTERS(A0, A1) \
|
||||
do { \
|
||||
A0 = _mm512_permutexvar_epi64(_mm512_setr_epi64(0, 1, 4, 5, 2, 3, 6, 7), A0); \
|
||||
A1 = _mm512_permutexvar_epi64(_mm512_setr_epi64(0, 1, 4, 5, 2, 3, 6, 7), A1); \
|
||||
SWAP_HALVES(A0, A1); \
|
||||
} while((void)0, 0)
|
||||
|
||||
#define BLAKE2_ROUND_1(A0, C0, B0, D0, A1, C1, B1, D1) \
|
||||
do { \
|
||||
SWAP_HALVES(A0, B0); \
|
||||
SWAP_HALVES(C0, D0); \
|
||||
SWAP_HALVES(A1, B1); \
|
||||
SWAP_HALVES(C1, D1); \
|
||||
BLAKE2_ROUND(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
SWAP_HALVES(A0, B0); \
|
||||
SWAP_HALVES(C0, D0); \
|
||||
SWAP_HALVES(A1, B1); \
|
||||
SWAP_HALVES(C1, D1); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define BLAKE2_ROUND_2(A0, A1, B0, B1, C0, C1, D0, D1) \
|
||||
do { \
|
||||
SWAP_QUARTERS(A0, A1); \
|
||||
SWAP_QUARTERS(B0, B1); \
|
||||
SWAP_QUARTERS(C0, C1); \
|
||||
SWAP_QUARTERS(D0, D1); \
|
||||
BLAKE2_ROUND(A0, B0, C0, D0, A1, B1, C1, D1); \
|
||||
UNSWAP_QUARTERS(A0, A1); \
|
||||
UNSWAP_QUARTERS(B0, B1); \
|
||||
UNSWAP_QUARTERS(C0, C1); \
|
||||
UNSWAP_QUARTERS(D0, D1); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#endif /* __AVX512F__ */
|
||||
#endif /* BLAKE_ROUND_MKA_OPT_H */
|
||||
56
algo/argon2/argon2d/blake2/blamka-round-ref.h
Normal file
56
algo/argon2/argon2d/blake2/blamka-round-ref.h
Normal file
@@ -0,0 +1,56 @@
|
||||
/*
|
||||
* Argon2 reference source code package - reference C implementations
|
||||
*
|
||||
* Copyright 2015
|
||||
* Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
|
||||
*
|
||||
* You may use this work under the terms of a Creative Commons CC0 1.0
|
||||
* License/Waiver or the Apache Public License 2.0, at your option. The terms of
|
||||
* these licenses can be found at:
|
||||
*
|
||||
* - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
|
||||
* - Apache 2.0 : http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* You should have received a copy of both of these licenses along with this
|
||||
* software. If not, they may be obtained at the above URLs.
|
||||
*/
|
||||
|
||||
#ifndef BLAKE_ROUND_MKA_H
|
||||
#define BLAKE_ROUND_MKA_H
|
||||
|
||||
#include "blake2.h"
|
||||
#include "blake2-impl.h"
|
||||
|
||||
/* designed by the Lyra PHC team */
|
||||
static BLAKE2_INLINE uint64_t fBlaMka(uint64_t x, uint64_t y) {
|
||||
const uint64_t m = UINT64_C(0xFFFFFFFF);
|
||||
const uint64_t xy = (x & m) * (y & m);
|
||||
return x + y + 2 * xy;
|
||||
}
|
||||
|
||||
#define G(a, b, c, d) \
|
||||
do { \
|
||||
a = fBlaMka(a, b); \
|
||||
d = rotr64(d ^ a, 32); \
|
||||
c = fBlaMka(c, d); \
|
||||
b = rotr64(b ^ c, 24); \
|
||||
a = fBlaMka(a, b); \
|
||||
d = rotr64(d ^ a, 16); \
|
||||
c = fBlaMka(c, d); \
|
||||
b = rotr64(b ^ c, 63); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#define BLAKE2_ROUND_NOMSG(v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11, \
|
||||
v12, v13, v14, v15) \
|
||||
do { \
|
||||
G(v0, v4, v8, v12); \
|
||||
G(v1, v5, v9, v13); \
|
||||
G(v2, v6, v10, v14); \
|
||||
G(v3, v7, v11, v15); \
|
||||
G(v0, v5, v10, v15); \
|
||||
G(v1, v6, v11, v12); \
|
||||
G(v2, v7, v8, v13); \
|
||||
G(v3, v4, v9, v14); \
|
||||
} while ((void)0, 0)
|
||||
|
||||
#endif
|
||||
@@ -15,7 +15,7 @@ void blakehash_4way(void *state, const void *input)
|
||||
memcpy( &ctx, &blake_4w_ctx, sizeof ctx );
|
||||
blake256r14_4way( &ctx, input + (64<<2), 16 );
|
||||
blake256r14_4way_close( &ctx, vhash );
|
||||
mm_deinterleave_4x32( state, state+32, state+64, state+96, vhash, 256 );
|
||||
mm128_deinterleave_4x32( state, state+32, state+64, state+96, vhash, 256 );
|
||||
}
|
||||
|
||||
int scanhash_blake_4way( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
@@ -37,7 +37,7 @@ int scanhash_blake_4way( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
|
||||
// we need big endian data...
|
||||
swab32_array( edata, pdata, 20 );
|
||||
mm_interleave_4x32( vdata, edata, edata, edata, edata, 640 );
|
||||
mm128_interleave_4x32( vdata, edata, edata, edata, edata, 640 );
|
||||
blake256r14_4way_init( &blake_4w_ctx );
|
||||
blake256r14_4way( &blake_4w_ctx, vdata, 64 );
|
||||
|
||||
|
||||
@@ -10,7 +10,7 @@ bool register_blake_algo( algo_gate_t* gate )
|
||||
gate->optimizations = AVX2_OPT;
|
||||
gate->get_max64 = (void*)&blake_get_max64;
|
||||
//#if defined (__AVX2__) && defined (FOUR_WAY)
|
||||
// gate->optimizations = SSE2_OPT | AVX_OPT | AVX2_OPT;
|
||||
// gate->optimizations = SSE2_OPT | AVX2_OPT;
|
||||
// gate->scanhash = (void*)&scanhash_blake_8way;
|
||||
// gate->hash = (void*)&blakehash_8way;
|
||||
#if defined(BLAKE_4WAY)
|
||||
|
||||
@@ -37,7 +37,7 @@
|
||||
#ifndef __BLAKE_HASH_4WAY__
|
||||
#define __BLAKE_HASH_4WAY__ 1
|
||||
|
||||
#ifdef __AVX__
|
||||
//#ifdef __SSE4_2__
|
||||
|
||||
#ifdef __cplusplus
|
||||
extern "C"{
|
||||
@@ -45,31 +45,34 @@ extern "C"{
|
||||
|
||||
#include <stddef.h>
|
||||
#include "algo/sha/sph_types.h"
|
||||
#include "avxdefs.h"
|
||||
#include "simd-utils.h"
|
||||
|
||||
#define SPH_SIZE_blake256 256
|
||||
|
||||
#define SPH_SIZE_blake512 512
|
||||
|
||||
// With AVX only Blake-256 4 way is available.
|
||||
// With SSE4.2 only Blake-256 4 way is available.
|
||||
// With AVX2 Blake-256 8way & Blake-512 4 way are also available.
|
||||
|
||||
// Blake-256 4 way
|
||||
|
||||
typedef struct {
|
||||
__m128i buf[16] __attribute__ ((aligned (64)));
|
||||
__m128i H[8];
|
||||
__m128i S[4];
|
||||
unsigned char buf[64<<2];
|
||||
uint32_t H[8<<2];
|
||||
uint32_t S[4<<2];
|
||||
// __m128i buf[16] __attribute__ ((aligned (64)));
|
||||
// __m128i H[8];
|
||||
// __m128i S[4];
|
||||
size_t ptr;
|
||||
sph_u32 T0, T1;
|
||||
uint32_t T0, T1;
|
||||
int rounds; // 14 for blake, 8 for blakecoin & vanilla
|
||||
} blake_4way_small_context;
|
||||
} blake_4way_small_context __attribute__ ((aligned (64)));
|
||||
|
||||
// Default 14 rounds
|
||||
typedef blake_4way_small_context blake256_4way_context;
|
||||
void blake256_4way_init(void *cc);
|
||||
void blake256_4way(void *cc, const void *data, size_t len);
|
||||
void blake256_4way_close(void *cc, void *dst);
|
||||
void blake256_4way_init(void *ctx);
|
||||
void blake256_4way(void *ctx, const void *data, size_t len);
|
||||
void blake256_4way_close(void *ctx, void *dst);
|
||||
|
||||
// 14 rounds, blake, decred
|
||||
typedef blake_4way_small_context blake256r14_4way_context;
|
||||
@@ -132,12 +135,10 @@ void blake512_4way_close(void *cc, void *dst);
|
||||
void blake512_4way_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
#endif
|
||||
#endif // AVX2
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif
|
||||
|
||||
#endif
|
||||
#endif // BLAKE_HASH_4WAY_H__
|
||||
|
||||
@@ -30,9 +30,10 @@
|
||||
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
|
||||
*/
|
||||
|
||||
#if defined (__AVX__)
|
||||
//#if defined (__SSE4_2__)
|
||||
|
||||
#include <stddef.h>
|
||||
#include <stdint.h>
|
||||
#include <string.h>
|
||||
#include <limits.h>
|
||||
|
||||
@@ -60,26 +61,12 @@ extern "C"{
|
||||
|
||||
// Blake-256
|
||||
|
||||
static const sph_u32 IV256[8] = {
|
||||
SPH_C32(0x6A09E667), SPH_C32(0xBB67AE85),
|
||||
SPH_C32(0x3C6EF372), SPH_C32(0xA54FF53A),
|
||||
SPH_C32(0x510E527F), SPH_C32(0x9B05688C),
|
||||
SPH_C32(0x1F83D9AB), SPH_C32(0x5BE0CD19)
|
||||
static const uint32_t IV256[8] =
|
||||
{
|
||||
0x6A09E667, 0xBB67AE85, 0x3C6EF372, 0xA54FF53A,
|
||||
0x510E527F, 0x9B05688C, 0x1F83D9AB, 0x5BE0CD19
|
||||
};
|
||||
|
||||
#if defined (__AVX2__)
|
||||
|
||||
// Blake-512
|
||||
|
||||
static const sph_u64 IV512[8] = {
|
||||
SPH_C64(0x6A09E667F3BCC908), SPH_C64(0xBB67AE8584CAA73B),
|
||||
SPH_C64(0x3C6EF372FE94F82B), SPH_C64(0xA54FF53A5F1D36F1),
|
||||
SPH_C64(0x510E527FADE682D1), SPH_C64(0x9B05688C2B3E6C1F),
|
||||
SPH_C64(0x1F83D9ABFB41BD6B), SPH_C64(0x5BE0CD19137E2179)
|
||||
};
|
||||
|
||||
#endif
|
||||
|
||||
#if SPH_COMPACT_BLAKE_32 || SPH_COMPACT_BLAKE_64
|
||||
|
||||
// Blake-256 4 & 8 way, Blake-512 4 way
|
||||
@@ -317,60 +304,19 @@ static const sph_u32 CS[16] = {
|
||||
|
||||
#endif
|
||||
|
||||
#if defined(__AVX2__)
|
||||
|
||||
// Blake-512 4 way
|
||||
|
||||
#define CBx(r, i) CBx_(Z ## r ## i)
|
||||
#define CBx_(n) CBx__(n)
|
||||
#define CBx__(n) CB ## n
|
||||
|
||||
#define CB0 SPH_C64(0x243F6A8885A308D3)
|
||||
#define CB1 SPH_C64(0x13198A2E03707344)
|
||||
#define CB2 SPH_C64(0xA4093822299F31D0)
|
||||
#define CB3 SPH_C64(0x082EFA98EC4E6C89)
|
||||
#define CB4 SPH_C64(0x452821E638D01377)
|
||||
#define CB5 SPH_C64(0xBE5466CF34E90C6C)
|
||||
#define CB6 SPH_C64(0xC0AC29B7C97C50DD)
|
||||
#define CB7 SPH_C64(0x3F84D5B5B5470917)
|
||||
#define CB8 SPH_C64(0x9216D5D98979FB1B)
|
||||
#define CB9 SPH_C64(0xD1310BA698DFB5AC)
|
||||
#define CBA SPH_C64(0x2FFD72DBD01ADFB7)
|
||||
#define CBB SPH_C64(0xB8E1AFED6A267E96)
|
||||
#define CBC SPH_C64(0xBA7C9045F12C7F99)
|
||||
#define CBD SPH_C64(0x24A19947B3916CF7)
|
||||
#define CBE SPH_C64(0x0801F2E2858EFC16)
|
||||
#define CBF SPH_C64(0x636920D871574E69)
|
||||
|
||||
#if SPH_COMPACT_BLAKE_64
|
||||
// not used
|
||||
static const sph_u64 CB[16] = {
|
||||
SPH_C64(0x243F6A8885A308D3), SPH_C64(0x13198A2E03707344),
|
||||
SPH_C64(0xA4093822299F31D0), SPH_C64(0x082EFA98EC4E6C89),
|
||||
SPH_C64(0x452821E638D01377), SPH_C64(0xBE5466CF34E90C6C),
|
||||
SPH_C64(0xC0AC29B7C97C50DD), SPH_C64(0x3F84D5B5B5470917),
|
||||
SPH_C64(0x9216D5D98979FB1B), SPH_C64(0xD1310BA698DFB5AC),
|
||||
SPH_C64(0x2FFD72DBD01ADFB7), SPH_C64(0xB8E1AFED6A267E96),
|
||||
SPH_C64(0xBA7C9045F12C7F99), SPH_C64(0x24A19947B3916CF7),
|
||||
SPH_C64(0x0801F2E2858EFC16), SPH_C64(0x636920D871574E69)
|
||||
};
|
||||
|
||||
#endif
|
||||
|
||||
#endif
|
||||
|
||||
#define GS_4WAY( m0, m1, c0, c1, a, b, c, d ) \
|
||||
do { \
|
||||
a = _mm_add_epi32( _mm_add_epi32( _mm_xor_si128( \
|
||||
_mm_set_epi32( c1, c1, c1, c1 ), m0 ), b ), a ); \
|
||||
d = mm_rotr_32( _mm_xor_si128( d, a ), 16 ); \
|
||||
d = mm128_ror_32( _mm_xor_si128( d, a ), 16 ); \
|
||||
c = _mm_add_epi32( c, d ); \
|
||||
b = mm_rotr_32( _mm_xor_si128( b, c ), 12 ); \
|
||||
b = mm128_ror_32( _mm_xor_si128( b, c ), 12 ); \
|
||||
a = _mm_add_epi32( _mm_add_epi32( _mm_xor_si128( \
|
||||
_mm_set_epi32( c0, c0, c0, c0 ), m1 ), b ), a ); \
|
||||
d = mm_rotr_32( _mm_xor_si128( d, a ), 8 ); \
|
||||
d = mm128_ror_32( _mm_xor_si128( d, a ), 8 ); \
|
||||
c = _mm_add_epi32( c, d ); \
|
||||
b = mm_rotr_32( _mm_xor_si128( b, c ), 7 ); \
|
||||
b = mm128_ror_32( _mm_xor_si128( b, c ), 7 ); \
|
||||
} while (0)
|
||||
|
||||
#if SPH_COMPACT_BLAKE_32
|
||||
@@ -411,125 +357,41 @@ do { \
|
||||
|
||||
#endif
|
||||
|
||||
#if defined (__AVX2__)
|
||||
|
||||
// Blake-256 8 way
|
||||
|
||||
#define GS_8WAY( m0, m1, c0, c1, a, b, c, d ) \
|
||||
do { \
|
||||
a = _mm256_add_epi32( _mm256_add_epi32( _mm256_xor_si256( \
|
||||
_mm256_set1_epi32( c1 ), m0 ), b ), a ); \
|
||||
d = mm256_rotr_32( _mm256_xor_si256( d, a ), 16 ); \
|
||||
c = _mm256_add_epi32( c, d ); \
|
||||
b = mm256_rotr_32( _mm256_xor_si256( b, c ), 12 ); \
|
||||
a = _mm256_add_epi32( _mm256_add_epi32( _mm256_xor_si256( \
|
||||
_mm256_set1_epi32( c0 ), m1 ), b ), a ); \
|
||||
d = mm256_rotr_32( _mm256_xor_si256( d, a ), 8 ); \
|
||||
c = _mm256_add_epi32( c, d ); \
|
||||
b = mm256_rotr_32( _mm256_xor_si256( b, c ), 7 ); \
|
||||
} while (0)
|
||||
|
||||
#define ROUND_S_8WAY(r) do { \
|
||||
GS_8WAY(Mx(r, 0), Mx(r, 1), CSx(r, 0), CSx(r, 1), V0, V4, V8, VC); \
|
||||
GS_8WAY(Mx(r, 2), Mx(r, 3), CSx(r, 2), CSx(r, 3), V1, V5, V9, VD); \
|
||||
GS_8WAY(Mx(r, 4), Mx(r, 5), CSx(r, 4), CSx(r, 5), V2, V6, VA, VE); \
|
||||
GS_8WAY(Mx(r, 6), Mx(r, 7), CSx(r, 6), CSx(r, 7), V3, V7, VB, VF); \
|
||||
GS_8WAY(Mx(r, 8), Mx(r, 9), CSx(r, 8), CSx(r, 9), V0, V5, VA, VF); \
|
||||
GS_8WAY(Mx(r, A), Mx(r, B), CSx(r, A), CSx(r, B), V1, V6, VB, VC); \
|
||||
GS_8WAY(Mx(r, C), Mx(r, D), CSx(r, C), CSx(r, D), V2, V7, V8, VD); \
|
||||
GS_8WAY(Mx(r, E), Mx(r, F), CSx(r, E), CSx(r, F), V3, V4, V9, VE); \
|
||||
} while (0)
|
||||
|
||||
// Blake-512 4 way
|
||||
|
||||
#define GB_4WAY(m0, m1, c0, c1, a, b, c, d) do { \
|
||||
a = _mm256_add_epi64( _mm256_add_epi64( _mm256_xor_si256( \
|
||||
_mm256_set_epi64x( c1, c1, c1, c1 ), m0 ), b ), a ); \
|
||||
d = mm256_rotr_64( _mm256_xor_si256( d, a ), 32 ); \
|
||||
c = _mm256_add_epi64( c, d ); \
|
||||
b = mm256_rotr_64( _mm256_xor_si256( b, c ), 25 ); \
|
||||
a = _mm256_add_epi64( _mm256_add_epi64( _mm256_xor_si256( \
|
||||
_mm256_set_epi64x( c0, c0, c0, c0 ), m1 ), b ), a ); \
|
||||
d = mm256_rotr_64( _mm256_xor_si256( d, a ), 16 ); \
|
||||
c = _mm256_add_epi64( c, d ); \
|
||||
b = mm256_rotr_64( _mm256_xor_si256( b, c ), 11 ); \
|
||||
} while (0)
|
||||
|
||||
#if SPH_COMPACT_BLAKE_64
|
||||
// not used
|
||||
#define ROUND_B_4WAY(r) do { \
|
||||
GB_4WAY(M[sigma[r][0x0]], M[sigma[r][0x1]], \
|
||||
CB[sigma[r][0x0]], CB[sigma[r][0x1]], V0, V4, V8, VC); \
|
||||
GB_4WAY(M[sigma[r][0x2]], M[sigma[r][0x3]], \
|
||||
CB[sigma[r][0x2]], CB[sigma[r][0x3]], V1, V5, V9, VD); \
|
||||
GB_4WAY(M[sigma[r][0x4]], M[sigma[r][0x5]], \
|
||||
CB[sigma[r][0x4]], CB[sigma[r][0x5]], V2, V6, VA, VE); \
|
||||
GB_4WAY(M[sigma[r][0x6]], M[sigma[r][0x7]], \
|
||||
CB[sigma[r][0x6]], CB[sigma[r][0x7]], V3, V7, VB, VF); \
|
||||
GB_4WAY(M[sigma[r][0x8]], M[sigma[r][0x9]], \
|
||||
CB[sigma[r][0x8]], CB[sigma[r][0x9]], V0, V5, VA, VF); \
|
||||
GB_4WAY(M[sigma[r][0xA]], M[sigma[r][0xB]], \
|
||||
CB[sigma[r][0xA]], CB[sigma[r][0xB]], V1, V6, VB, VC); \
|
||||
GB_4WAY(M[sigma[r][0xC]], M[sigma[r][0xD]], \
|
||||
CB[sigma[r][0xC]], CB[sigma[r][0xD]], V2, V7, V8, VD); \
|
||||
GB_4WAY(M[sigma[r][0xE]], M[sigma[r][0xF]], \
|
||||
CB[sigma[r][0xE]], CB[sigma[r][0xF]], V3, V4, V9, VE); \
|
||||
} while (0)
|
||||
|
||||
#else
|
||||
//current_impl
|
||||
#define ROUND_B_4WAY(r) do { \
|
||||
GB_4WAY(Mx(r, 0), Mx(r, 1), CBx(r, 0), CBx(r, 1), V0, V4, V8, VC); \
|
||||
GB_4WAY(Mx(r, 2), Mx(r, 3), CBx(r, 2), CBx(r, 3), V1, V5, V9, VD); \
|
||||
GB_4WAY(Mx(r, 4), Mx(r, 5), CBx(r, 4), CBx(r, 5), V2, V6, VA, VE); \
|
||||
GB_4WAY(Mx(r, 6), Mx(r, 7), CBx(r, 6), CBx(r, 7), V3, V7, VB, VF); \
|
||||
GB_4WAY(Mx(r, 8), Mx(r, 9), CBx(r, 8), CBx(r, 9), V0, V5, VA, VF); \
|
||||
GB_4WAY(Mx(r, A), Mx(r, B), CBx(r, A), CBx(r, B), V1, V6, VB, VC); \
|
||||
GB_4WAY(Mx(r, C), Mx(r, D), CBx(r, C), CBx(r, D), V2, V7, V8, VD); \
|
||||
GB_4WAY(Mx(r, E), Mx(r, F), CBx(r, E), CBx(r, F), V3, V4, V9, VE); \
|
||||
} while (0)
|
||||
|
||||
#endif
|
||||
|
||||
#endif
|
||||
|
||||
// Blake-256 4 way
|
||||
|
||||
#define DECL_STATE32_4WAY \
|
||||
__m128i H0, H1, H2, H3, H4, H5, H6, H7; \
|
||||
__m128i S0, S1, S2, S3; \
|
||||
sph_u32 T0, T1;
|
||||
uint32_t T0, T1;
|
||||
|
||||
#define READ_STATE32_4WAY(state) do { \
|
||||
H0 = (state)->H[0]; \
|
||||
H1 = (state)->H[1]; \
|
||||
H2 = (state)->H[2]; \
|
||||
H3 = (state)->H[3]; \
|
||||
H4 = (state)->H[4]; \
|
||||
H5 = (state)->H[5]; \
|
||||
H6 = (state)->H[6]; \
|
||||
H7 = (state)->H[7]; \
|
||||
S0 = (state)->S[0]; \
|
||||
S1 = (state)->S[1]; \
|
||||
S2 = (state)->S[2]; \
|
||||
S3 = (state)->S[3]; \
|
||||
H0 = casti_m128i( state->H, 0 ); \
|
||||
H1 = casti_m128i( state->H, 1 ); \
|
||||
H2 = casti_m128i( state->H, 2 ); \
|
||||
H3 = casti_m128i( state->H, 3 ); \
|
||||
H4 = casti_m128i( state->H, 4 ); \
|
||||
H5 = casti_m128i( state->H, 5 ); \
|
||||
H6 = casti_m128i( state->H, 6 ); \
|
||||
H7 = casti_m128i( state->H, 7 ); \
|
||||
S0 = casti_m128i( state->S, 0 ); \
|
||||
S1 = casti_m128i( state->S, 1 ); \
|
||||
S2 = casti_m128i( state->S, 2 ); \
|
||||
S3 = casti_m128i( state->S, 3 ); \
|
||||
T0 = (state)->T0; \
|
||||
T1 = (state)->T1; \
|
||||
} while (0)
|
||||
|
||||
#define WRITE_STATE32_4WAY(state) do { \
|
||||
(state)->H[0] = H0; \
|
||||
(state)->H[1] = H1; \
|
||||
(state)->H[2] = H2; \
|
||||
(state)->H[3] = H3; \
|
||||
(state)->H[4] = H4; \
|
||||
(state)->H[5] = H5; \
|
||||
(state)->H[6] = H6; \
|
||||
(state)->H[7] = H7; \
|
||||
(state)->S[0] = S0; \
|
||||
(state)->S[1] = S1; \
|
||||
(state)->S[2] = S2; \
|
||||
(state)->S[3] = S3; \
|
||||
casti_m128i( state->H, 0 ) = H0; \
|
||||
casti_m128i( state->H, 1 ) = H1; \
|
||||
casti_m128i( state->H, 2 ) = H2; \
|
||||
casti_m128i( state->H, 3 ) = H3; \
|
||||
casti_m128i( state->H, 4 ) = H4; \
|
||||
casti_m128i( state->H, 5 ) = H5; \
|
||||
casti_m128i( state->H, 6 ) = H6; \
|
||||
casti_m128i( state->H, 7 ) = H7; \
|
||||
casti_m128i( state->S, 0 ) = S0; \
|
||||
casti_m128i( state->S, 1 ) = S1; \
|
||||
casti_m128i( state->S, 2 ) = S2; \
|
||||
casti_m128i( state->S, 3 ) = S3; \
|
||||
(state)->T0 = T0; \
|
||||
(state)->T1 = T1; \
|
||||
} while (0)
|
||||
@@ -562,22 +424,22 @@ do { \
|
||||
, _mm_set_epi32( CS6, CS6, CS6, CS6 ) ); \
|
||||
VF = _mm_xor_si128( _mm_set_epi32( T1, T1, T1, T1 ), \
|
||||
_mm_set_epi32( CS7, CS7, CS7, CS7 ) ); \
|
||||
M[0x0] = mm_bswap_32( *(buf + 0) ); \
|
||||
M[0x1] = mm_bswap_32( *(buf + 1) ); \
|
||||
M[0x2] = mm_bswap_32( *(buf + 2) ); \
|
||||
M[0x3] = mm_bswap_32( *(buf + 3) ); \
|
||||
M[0x4] = mm_bswap_32( *(buf + 4) ); \
|
||||
M[0x5] = mm_bswap_32( *(buf + 5) ); \
|
||||
M[0x6] = mm_bswap_32( *(buf + 6) ); \
|
||||
M[0x7] = mm_bswap_32( *(buf + 7) ); \
|
||||
M[0x8] = mm_bswap_32( *(buf + 8) ); \
|
||||
M[0x9] = mm_bswap_32( *(buf + 9) ); \
|
||||
M[0xA] = mm_bswap_32( *(buf + 10) ); \
|
||||
M[0xB] = mm_bswap_32( *(buf + 11) ); \
|
||||
M[0xC] = mm_bswap_32( *(buf + 12) ); \
|
||||
M[0xD] = mm_bswap_32( *(buf + 13) ); \
|
||||
M[0xE] = mm_bswap_32( *(buf + 14) ); \
|
||||
M[0xF] = mm_bswap_32( *(buf + 15) ); \
|
||||
M[0x0] = mm128_bswap_32( *(buf + 0) ); \
|
||||
M[0x1] = mm128_bswap_32( *(buf + 1) ); \
|
||||
M[0x2] = mm128_bswap_32( *(buf + 2) ); \
|
||||
M[0x3] = mm128_bswap_32( *(buf + 3) ); \
|
||||
M[0x4] = mm128_bswap_32( *(buf + 4) ); \
|
||||
M[0x5] = mm128_bswap_32( *(buf + 5) ); \
|
||||
M[0x6] = mm128_bswap_32( *(buf + 6) ); \
|
||||
M[0x7] = mm128_bswap_32( *(buf + 7) ); \
|
||||
M[0x8] = mm128_bswap_32( *(buf + 8) ); \
|
||||
M[0x9] = mm128_bswap_32( *(buf + 9) ); \
|
||||
M[0xA] = mm128_bswap_32( *(buf + 10) ); \
|
||||
M[0xB] = mm128_bswap_32( *(buf + 11) ); \
|
||||
M[0xC] = mm128_bswap_32( *(buf + 12) ); \
|
||||
M[0xD] = mm128_bswap_32( *(buf + 13) ); \
|
||||
M[0xE] = mm128_bswap_32( *(buf + 14) ); \
|
||||
M[0xF] = mm128_bswap_32( *(buf + 15) ); \
|
||||
for (r = 0; r < rounds; r ++) \
|
||||
ROUND_S_4WAY(r); \
|
||||
H0 = _mm_xor_si128( _mm_xor_si128( \
|
||||
@@ -616,30 +478,30 @@ do { \
|
||||
V5 = H5; \
|
||||
V6 = H6; \
|
||||
V7 = H7; \
|
||||
V8 = _mm_xor_si128( S0, _mm_set_epi32( CS0, CS0, CS0, CS0 ) ); \
|
||||
V9 = _mm_xor_si128( S1, _mm_set_epi32( CS1, CS1, CS1, CS1 ) ); \
|
||||
VA = _mm_xor_si128( S2, _mm_set_epi32( CS2, CS2, CS2, CS2 ) ); \
|
||||
VB = _mm_xor_si128( S3, _mm_set_epi32( CS3, CS3, CS3, CS3 ) ); \
|
||||
V8 = _mm_xor_si128( S0, _mm_set1_epi32( CS0 ) ); \
|
||||
V9 = _mm_xor_si128( S1, _mm_set1_epi32( CS1 ) ); \
|
||||
VA = _mm_xor_si128( S2, _mm_set1_epi32( CS2 ) ); \
|
||||
VB = _mm_xor_si128( S3, _mm_set1_epi32( CS3 ) ); \
|
||||
VC = _mm_xor_si128( _mm_set1_epi32( T0 ), _mm_set1_epi32( CS4 ) ); \
|
||||
VD = _mm_xor_si128( _mm_set1_epi32( T0 ), _mm_set1_epi32( CS5 ) ); \
|
||||
VE = _mm_xor_si128( _mm_set1_epi32( T1 ), _mm_set1_epi32( CS6 ) ); \
|
||||
VF = _mm_xor_si128( _mm_set1_epi32( T1 ), _mm_set1_epi32( CS7 ) ); \
|
||||
M0 = mm_bswap_32( * buf ); \
|
||||
M1 = mm_bswap_32( *(buf+1) ); \
|
||||
M2 = mm_bswap_32( *(buf+2) ); \
|
||||
M3 = mm_bswap_32( *(buf+3) ); \
|
||||
M4 = mm_bswap_32( *(buf+4) ); \
|
||||
M5 = mm_bswap_32( *(buf+5) ); \
|
||||
M6 = mm_bswap_32( *(buf+6) ); \
|
||||
M7 = mm_bswap_32( *(buf+7) ); \
|
||||
M8 = mm_bswap_32( *(buf+8) ); \
|
||||
M9 = mm_bswap_32( *(buf+9) ); \
|
||||
MA = mm_bswap_32( *(buf+10) ); \
|
||||
MB = mm_bswap_32( *(buf+11) ); \
|
||||
MC = mm_bswap_32( *(buf+12) ); \
|
||||
MD = mm_bswap_32( *(buf+13) ); \
|
||||
ME = mm_bswap_32( *(buf+14) ); \
|
||||
MF = mm_bswap_32( *(buf+15) ); \
|
||||
M0 = mm128_bswap_32( buf[ 0] ); \
|
||||
M1 = mm128_bswap_32( buf[ 1] ); \
|
||||
M2 = mm128_bswap_32( buf[ 2] ); \
|
||||
M3 = mm128_bswap_32( buf[ 3] ); \
|
||||
M4 = mm128_bswap_32( buf[ 4] ); \
|
||||
M5 = mm128_bswap_32( buf[ 5] ); \
|
||||
M6 = mm128_bswap_32( buf[ 6] ); \
|
||||
M7 = mm128_bswap_32( buf[ 7] ); \
|
||||
M8 = mm128_bswap_32( buf[ 8] ); \
|
||||
M9 = mm128_bswap_32( buf[ 9] ); \
|
||||
MA = mm128_bswap_32( buf[10] ); \
|
||||
MB = mm128_bswap_32( buf[11] ); \
|
||||
MC = mm128_bswap_32( buf[12] ); \
|
||||
MD = mm128_bswap_32( buf[13] ); \
|
||||
ME = mm128_bswap_32( buf[14] ); \
|
||||
MF = mm128_bswap_32( buf[15] ); \
|
||||
ROUND_S_4WAY(0); \
|
||||
ROUND_S_4WAY(1); \
|
||||
ROUND_S_4WAY(2); \
|
||||
@@ -673,6 +535,31 @@ do { \
|
||||
|
||||
// Blake-256 8 way
|
||||
|
||||
#define GS_8WAY( m0, m1, c0, c1, a, b, c, d ) \
|
||||
do { \
|
||||
a = _mm256_add_epi32( _mm256_add_epi32( _mm256_xor_si256( \
|
||||
_mm256_set1_epi32( c1 ), m0 ), b ), a ); \
|
||||
d = mm256_ror_32( _mm256_xor_si256( d, a ), 16 ); \
|
||||
c = _mm256_add_epi32( c, d ); \
|
||||
b = mm256_ror_32( _mm256_xor_si256( b, c ), 12 ); \
|
||||
a = _mm256_add_epi32( _mm256_add_epi32( _mm256_xor_si256( \
|
||||
_mm256_set1_epi32( c0 ), m1 ), b ), a ); \
|
||||
d = mm256_ror_32( _mm256_xor_si256( d, a ), 8 ); \
|
||||
c = _mm256_add_epi32( c, d ); \
|
||||
b = mm256_ror_32( _mm256_xor_si256( b, c ), 7 ); \
|
||||
} while (0)
|
||||
|
||||
#define ROUND_S_8WAY(r) do { \
|
||||
GS_8WAY(Mx(r, 0), Mx(r, 1), CSx(r, 0), CSx(r, 1), V0, V4, V8, VC); \
|
||||
GS_8WAY(Mx(r, 2), Mx(r, 3), CSx(r, 2), CSx(r, 3), V1, V5, V9, VD); \
|
||||
GS_8WAY(Mx(r, 4), Mx(r, 5), CSx(r, 4), CSx(r, 5), V2, V6, VA, VE); \
|
||||
GS_8WAY(Mx(r, 6), Mx(r, 7), CSx(r, 6), CSx(r, 7), V3, V7, VB, VF); \
|
||||
GS_8WAY(Mx(r, 8), Mx(r, 9), CSx(r, 8), CSx(r, 9), V0, V5, VA, VF); \
|
||||
GS_8WAY(Mx(r, A), Mx(r, B), CSx(r, A), CSx(r, B), V1, V6, VB, VC); \
|
||||
GS_8WAY(Mx(r, C), Mx(r, D), CSx(r, C), CSx(r, D), V2, V7, V8, VD); \
|
||||
GS_8WAY(Mx(r, E), Mx(r, F), CSx(r, E), CSx(r, F), V3, V4, V9, VE); \
|
||||
} while (0)
|
||||
|
||||
#define DECL_STATE32_8WAY \
|
||||
__m256i H0, H1, H2, H3, H4, H5, H6, H7; \
|
||||
__m256i S0, S1, S2, S3; \
|
||||
@@ -787,312 +674,136 @@ do { \
|
||||
S3 ), H7 ); \
|
||||
} while (0)
|
||||
|
||||
// Blake-512 4 way
|
||||
|
||||
#define DECL_STATE64_4WAY \
|
||||
__m256i H0, H1, H2, H3, H4, H5, H6, H7; \
|
||||
__m256i S0, S1, S2, S3; \
|
||||
sph_u64 T0, T1;
|
||||
|
||||
#define READ_STATE64_4WAY(state) do { \
|
||||
H0 = (state)->H[0]; \
|
||||
H1 = (state)->H[1]; \
|
||||
H2 = (state)->H[2]; \
|
||||
H3 = (state)->H[3]; \
|
||||
H4 = (state)->H[4]; \
|
||||
H5 = (state)->H[5]; \
|
||||
H6 = (state)->H[6]; \
|
||||
H7 = (state)->H[7]; \
|
||||
S0 = (state)->S[0]; \
|
||||
S1 = (state)->S[1]; \
|
||||
S2 = (state)->S[2]; \
|
||||
S3 = (state)->S[3]; \
|
||||
T0 = (state)->T0; \
|
||||
T1 = (state)->T1; \
|
||||
} while (0)
|
||||
|
||||
#define WRITE_STATE64_4WAY(state) do { \
|
||||
(state)->H[0] = H0; \
|
||||
(state)->H[1] = H1; \
|
||||
(state)->H[2] = H2; \
|
||||
(state)->H[3] = H3; \
|
||||
(state)->H[4] = H4; \
|
||||
(state)->H[5] = H5; \
|
||||
(state)->H[6] = H6; \
|
||||
(state)->H[7] = H7; \
|
||||
(state)->S[0] = S0; \
|
||||
(state)->S[1] = S1; \
|
||||
(state)->S[2] = S2; \
|
||||
(state)->S[3] = S3; \
|
||||
(state)->T0 = T0; \
|
||||
(state)->T1 = T1; \
|
||||
} while (0)
|
||||
|
||||
#if SPH_COMPACT_BLAKE_64
|
||||
|
||||
// not used
|
||||
#define COMPRESS64_4WAY do { \
|
||||
__m256i M[16]; \
|
||||
__m256i V0, V1, V2, V3, V4, V5, V6, V7; \
|
||||
__m256i V8, V9, VA, VB, VC, VD, VE, VF; \
|
||||
unsigned r; \
|
||||
V0 = H0; \
|
||||
V1 = H1; \
|
||||
V2 = H2; \
|
||||
V3 = H3; \
|
||||
V4 = H4; \
|
||||
V5 = H5; \
|
||||
V6 = H6; \
|
||||
V7 = H7; \
|
||||
V8 = _mm256_xor_si256( S0, _mm256_set_epi64x( CB0, CB0, CB0, CB0 ) ); \
|
||||
V9 = _mm256_xor_si256( S1, _mm256_set_epi64x( CB1, CB1, CB1, CB1 ) ); \
|
||||
VA = _mm256_xor_si256( S2, _mm256_set_epi64x( CB2, CB2, CB2, CB2 ) ); \
|
||||
VB = _mm256_xor_si256( S3, _mm256_set_epi64x( CB3, CB3, CB3, CB3 ) ); \
|
||||
VC = _mm256_xor_si256( _mm256_set_epi64x( T0, T0, T0, T0 ), \
|
||||
_mm256_set_epi64x( CB4, CB4, CB4, CB4 ) ); \
|
||||
VD = _mm256_xor_si256( _mm256_set_epi64x( T0, T0, T0, T0 ), \
|
||||
_mm256_set_epi64x( CB5, CB5, CB5, CB5 ) ); \
|
||||
VE = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
|
||||
_mm256_set_epi64x( CB6, CB6, CB6, CB6 ) ); \
|
||||
VF = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
|
||||
_mm256_set_epi64x( CB7, CB7, CB7, CB7 ) ); \
|
||||
M[0x0] = mm256_bswap_64( *(buf+0) ); \
|
||||
M[0x1] = mm256_bswap_64( *(buf+1) ); \
|
||||
M[0x2] = mm256_bswap_64( *(buf+2) ); \
|
||||
M[0x3] = mm256_bswap_64( *(buf+3) ); \
|
||||
M[0x4] = mm256_bswap_64( *(buf+4) ); \
|
||||
M[0x5] = mm256_bswap_64( *(buf+5) ); \
|
||||
M[0x6] = mm256_bswap_64( *(buf+6) ); \
|
||||
M[0x7] = mm256_bswap_64( *(buf+7) ); \
|
||||
M[0x8] = mm256_bswap_64( *(buf+8) ); \
|
||||
M[0x9] = mm256_bswap_64( *(buf+9) ); \
|
||||
M[0xA] = mm256_bswap_64( *(buf+10) ); \
|
||||
M[0xB] = mm256_bswap_64( *(buf+11) ); \
|
||||
M[0xC] = mm256_bswap_64( *(buf+12) ); \
|
||||
M[0xD] = mm256_bswap_64( *(buf+13) ); \
|
||||
M[0xE] = mm256_bswap_64( *(buf+14) ); \
|
||||
M[0xF] = mm256_bswap_64( *(buf+15) ); \
|
||||
for (r = 0; r < 16; r ++) \
|
||||
ROUND_B_4WAY(r); \
|
||||
H0 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S0, V0 ), V8 ), H0 ); \
|
||||
H1 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S1, V1 ), V9 ), H1 ); \
|
||||
H2 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S2, V2 ), VA ), H2 ); \
|
||||
H3 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S3, V3 ), VB ), H3 ); \
|
||||
H4 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S0, V4 ), VC ), H4 ); \
|
||||
H5 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S1, V5 ), VD ), H5 ); \
|
||||
H6 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S2, V6 ), VE ), H6 ); \
|
||||
H7 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S3, V7 ), VF ), H7 ); \
|
||||
} while (0)
|
||||
|
||||
#else
|
||||
|
||||
//current impl
|
||||
|
||||
#define COMPRESS64_4WAY do { \
|
||||
__m256i M0, M1, M2, M3, M4, M5, M6, M7; \
|
||||
__m256i M8, M9, MA, MB, MC, MD, ME, MF; \
|
||||
__m256i V0, V1, V2, V3, V4, V5, V6, V7; \
|
||||
__m256i V8, V9, VA, VB, VC, VD, VE, VF; \
|
||||
V0 = H0; \
|
||||
V1 = H1; \
|
||||
V2 = H2; \
|
||||
V3 = H3; \
|
||||
V4 = H4; \
|
||||
V5 = H5; \
|
||||
V6 = H6; \
|
||||
V7 = H7; \
|
||||
V8 = _mm256_xor_si256( S0, _mm256_set_epi64x( CB0, CB0, CB0, CB0 ) ); \
|
||||
V9 = _mm256_xor_si256( S1, _mm256_set_epi64x( CB1, CB1, CB1, CB1 ) ); \
|
||||
VA = _mm256_xor_si256( S2, _mm256_set_epi64x( CB2, CB2, CB2, CB2 ) ); \
|
||||
VB = _mm256_xor_si256( S3, _mm256_set_epi64x( CB3, CB3, CB3, CB3 ) ); \
|
||||
VC = _mm256_xor_si256( _mm256_set_epi64x( T0, T0, T0, T0 ), \
|
||||
_mm256_set_epi64x( CB4, CB4, CB4, CB4 ) ); \
|
||||
VD = _mm256_xor_si256( _mm256_set_epi64x( T0, T0, T0, T0 ), \
|
||||
_mm256_set_epi64x( CB5, CB5, CB5, CB5 ) ); \
|
||||
VE = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
|
||||
_mm256_set_epi64x( CB6, CB6, CB6, CB6 ) ); \
|
||||
VF = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
|
||||
_mm256_set_epi64x( CB7, CB7, CB7, CB7 ) ); \
|
||||
M0 = mm256_bswap_64( *(buf + 0) ); \
|
||||
M1 = mm256_bswap_64( *(buf + 1) ); \
|
||||
M2 = mm256_bswap_64( *(buf + 2) ); \
|
||||
M3 = mm256_bswap_64( *(buf + 3) ); \
|
||||
M4 = mm256_bswap_64( *(buf + 4) ); \
|
||||
M5 = mm256_bswap_64( *(buf + 5) ); \
|
||||
M6 = mm256_bswap_64( *(buf + 6) ); \
|
||||
M7 = mm256_bswap_64( *(buf + 7) ); \
|
||||
M8 = mm256_bswap_64( *(buf + 8) ); \
|
||||
M9 = mm256_bswap_64( *(buf + 9) ); \
|
||||
MA = mm256_bswap_64( *(buf + 10) ); \
|
||||
MB = mm256_bswap_64( *(buf + 11) ); \
|
||||
MC = mm256_bswap_64( *(buf + 12) ); \
|
||||
MD = mm256_bswap_64( *(buf + 13) ); \
|
||||
ME = mm256_bswap_64( *(buf + 14) ); \
|
||||
MF = mm256_bswap_64( *(buf + 15) ); \
|
||||
ROUND_B_4WAY(0); \
|
||||
ROUND_B_4WAY(1); \
|
||||
ROUND_B_4WAY(2); \
|
||||
ROUND_B_4WAY(3); \
|
||||
ROUND_B_4WAY(4); \
|
||||
ROUND_B_4WAY(5); \
|
||||
ROUND_B_4WAY(6); \
|
||||
ROUND_B_4WAY(7); \
|
||||
ROUND_B_4WAY(8); \
|
||||
ROUND_B_4WAY(9); \
|
||||
ROUND_B_4WAY(0); \
|
||||
ROUND_B_4WAY(1); \
|
||||
ROUND_B_4WAY(2); \
|
||||
ROUND_B_4WAY(3); \
|
||||
ROUND_B_4WAY(4); \
|
||||
ROUND_B_4WAY(5); \
|
||||
H0 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S0, V0 ), V8 ), H0 ); \
|
||||
H1 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S1, V1 ), V9 ), H1 ); \
|
||||
H2 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S2, V2 ), VA ), H2 ); \
|
||||
H3 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S3, V3 ), VB ), H3 ); \
|
||||
H4 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S0, V4 ), VC ), H4 ); \
|
||||
H5 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S1, V5 ), VD ), H5 ); \
|
||||
H6 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S2, V6 ), VE ), H6 ); \
|
||||
H7 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S3, V7 ), VF ), H7 ); \
|
||||
} while (0)
|
||||
|
||||
#endif
|
||||
|
||||
#endif
|
||||
|
||||
// Blake-256 4 way
|
||||
|
||||
static const sph_u32 salt_zero_4way_small[4] = { 0, 0, 0, 0 };
|
||||
static const uint32_t salt_zero_4way_small[4] = { 0, 0, 0, 0 };
|
||||
|
||||
static void
|
||||
blake32_4way_init( blake_4way_small_context *sc, const sph_u32 *iv,
|
||||
const sph_u32 *salt, int rounds )
|
||||
blake32_4way_init( blake_4way_small_context *ctx, const uint32_t *iv,
|
||||
const uint32_t *salt, int rounds )
|
||||
{
|
||||
int i;
|
||||
for ( i = 0; i < 8; i++ )
|
||||
sc->H[i] = _mm_set1_epi32( iv[i] );
|
||||
for ( i = 0; i < 4; i++ )
|
||||
sc->S[i] = _mm_set1_epi32( salt[i] );
|
||||
sc->T0 = sc->T1 = 0;
|
||||
sc->ptr = 0;
|
||||
sc->rounds = rounds;
|
||||
casti_m128i( ctx->H, 0 ) = _mm_set1_epi32( iv[0] );
|
||||
casti_m128i( ctx->H, 1 ) = _mm_set1_epi32( iv[1] );
|
||||
casti_m128i( ctx->H, 2 ) = _mm_set1_epi32( iv[2] );
|
||||
casti_m128i( ctx->H, 3 ) = _mm_set1_epi32( iv[3] );
|
||||
casti_m128i( ctx->H, 4 ) = _mm_set1_epi32( iv[4] );
|
||||
casti_m128i( ctx->H, 5 ) = _mm_set1_epi32( iv[5] );
|
||||
casti_m128i( ctx->H, 6 ) = _mm_set1_epi32( iv[6] );
|
||||
casti_m128i( ctx->H, 7 ) = _mm_set1_epi32( iv[7] );
|
||||
|
||||
casti_m128i( ctx->S, 0 ) = m128_zero;
|
||||
casti_m128i( ctx->S, 1 ) = m128_zero;
|
||||
casti_m128i( ctx->S, 2 ) = m128_zero;
|
||||
casti_m128i( ctx->S, 3 ) = m128_zero;
|
||||
/*
|
||||
sc->S[0] = _mm_set1_epi32( salt[0] );
|
||||
sc->S[1] = _mm_set1_epi32( salt[1] );
|
||||
sc->S[2] = _mm_set1_epi32( salt[2] );
|
||||
sc->S[3] = _mm_set1_epi32( salt[3] );
|
||||
*/
|
||||
ctx->T0 = ctx->T1 = 0;
|
||||
ctx->ptr = 0;
|
||||
ctx->rounds = rounds;
|
||||
}
|
||||
|
||||
static void
|
||||
blake32_4way( blake_4way_small_context *sc, const void *data, size_t len )
|
||||
blake32_4way( blake_4way_small_context *ctx, const void *data, size_t len )
|
||||
{
|
||||
__m128i *vdata = (__m128i*)data;
|
||||
__m128i *buf;
|
||||
size_t ptr;
|
||||
const int buf_size = 64; // number of elements, sizeof/4
|
||||
__m128i *buf = (__m128i*)ctx->buf;
|
||||
size_t bptr = ctx->ptr<<2;
|
||||
size_t vptr = ctx->ptr >> 2;
|
||||
size_t blen = len << 2;
|
||||
DECL_STATE32_4WAY
|
||||
buf = sc->buf;
|
||||
ptr = sc->ptr;
|
||||
if ( len < buf_size - ptr )
|
||||
|
||||
if ( blen < (sizeof ctx->buf) - bptr )
|
||||
{
|
||||
memcpy_128( buf + (ptr>>2), vdata, len>>2 );
|
||||
ptr += len;
|
||||
sc->ptr = ptr;
|
||||
memcpy( buf + vptr, data, (sizeof ctx->buf) - bptr );
|
||||
bptr += blen;
|
||||
ctx->ptr = bptr>>2;
|
||||
return;
|
||||
}
|
||||
|
||||
READ_STATE32_4WAY(sc);
|
||||
while ( len > 0 )
|
||||
READ_STATE32_4WAY( ctx );
|
||||
while ( blen > 0 )
|
||||
{
|
||||
size_t clen;
|
||||
size_t clen = ( sizeof ctx->buf ) - bptr;
|
||||
|
||||
clen = buf_size - ptr;
|
||||
if ( clen > len )
|
||||
clen = len;
|
||||
memcpy_128( buf + (ptr>>2), vdata, clen>>2 );
|
||||
ptr += clen;
|
||||
vdata += (clen>>2);
|
||||
len -= clen;
|
||||
if ( ptr == buf_size )
|
||||
if ( clen > blen )
|
||||
clen = blen;
|
||||
memcpy( buf + vptr, data, clen );
|
||||
bptr += clen;
|
||||
data = (const unsigned char *)data + clen;
|
||||
blen -= clen;
|
||||
if ( bptr == ( sizeof ctx->buf ) )
|
||||
{
|
||||
if ( ( T0 = SPH_T32(T0 + 512) ) < 512 )
|
||||
T1 = SPH_T32(T1 + 1);
|
||||
COMPRESS32_4WAY( sc->rounds );
|
||||
ptr = 0;
|
||||
if ( ( T0 = T0 + 512 ) < 512 )
|
||||
T1 = T1 + 1;
|
||||
COMPRESS32_4WAY( ctx->rounds );
|
||||
bptr = 0;
|
||||
}
|
||||
}
|
||||
WRITE_STATE32_4WAY(sc);
|
||||
sc->ptr = ptr;
|
||||
WRITE_STATE32_4WAY( ctx );
|
||||
ctx->ptr = bptr>>2;
|
||||
}
|
||||
|
||||
static void
|
||||
blake32_4way_close( blake_4way_small_context *sc, unsigned ub, unsigned n,
|
||||
blake32_4way_close( blake_4way_small_context *ctx, unsigned ub, unsigned n,
|
||||
void *dst, size_t out_size_w32 )
|
||||
{
|
||||
// union {
|
||||
__m128i buf[16];
|
||||
// sph_u32 dummy;
|
||||
// } u;
|
||||
size_t ptr, k;
|
||||
unsigned bit_len;
|
||||
sph_u32 th, tl;
|
||||
__m128i *out;
|
||||
|
||||
ptr = sc->ptr;
|
||||
bit_len = ((unsigned)ptr << 3);
|
||||
buf[ptr>>2] = _mm_set1_epi32( 0x80 );
|
||||
tl = sc->T0 + bit_len;
|
||||
th = sc->T1;
|
||||
__m128i buf[16] __attribute__ ((aligned (64)));
|
||||
size_t ptr = ctx->ptr;
|
||||
size_t vptr = ctx->ptr>>2;
|
||||
unsigned bit_len = ( (unsigned)ptr << 3 );
|
||||
uint32_t tl = ctx->T0 + bit_len;
|
||||
uint32_t th = ctx->T1;
|
||||
|
||||
if ( ptr == 0 )
|
||||
{
|
||||
sc->T0 = SPH_C32(0xFFFFFE00UL);
|
||||
sc->T1 = SPH_C32(0xFFFFFFFFUL);
|
||||
ctx->T0 = 0xFFFFFE00UL;
|
||||
ctx->T1 = 0xFFFFFFFFUL;
|
||||
}
|
||||
else if ( sc->T0 == 0 )
|
||||
else if ( ctx->T0 == 0 )
|
||||
{
|
||||
sc->T0 = SPH_C32(0xFFFFFE00UL) + bit_len;
|
||||
sc->T1 = SPH_T32(sc->T1 - 1);
|
||||
ctx->T0 = 0xFFFFFE00UL + bit_len;
|
||||
ctx->T1 = ctx->T1 - 1;
|
||||
}
|
||||
else
|
||||
sc->T0 -= 512 - bit_len;
|
||||
ctx->T0 -= 512 - bit_len;
|
||||
|
||||
if ( ptr <= 52 )
|
||||
buf[vptr] = _mm_set1_epi32( 0x80 );
|
||||
|
||||
if ( vptr < 12 )
|
||||
{
|
||||
memset_zero_128( buf + (ptr>>2) + 1, (52 - ptr) >> 2 );
|
||||
if (out_size_w32 == 8)
|
||||
buf[52>>2] = _mm_or_si128( buf[52>>2],
|
||||
_mm_set1_epi32( 0x01000000UL ) );
|
||||
*(buf+(56>>2)) = mm_bswap_32( _mm_set1_epi32( th ) );
|
||||
*(buf+(60>>2)) = mm_bswap_32( _mm_set1_epi32( tl ) );
|
||||
blake32_4way( sc, buf + (ptr>>2), 64 - ptr );
|
||||
memset_zero_128( buf + vptr + 1, 13 - vptr );
|
||||
buf[ 13 ] = _mm_or_si128( buf[ 13 ], _mm_set1_epi32( 0x01000000UL ) );
|
||||
buf[ 14 ] = mm128_bswap_32( _mm_set1_epi32( th ) );
|
||||
buf[ 15 ] = mm128_bswap_32( _mm_set1_epi32( tl ) );
|
||||
blake32_4way( ctx, buf + vptr, 64 - ptr );
|
||||
}
|
||||
else
|
||||
{
|
||||
memset_zero_128( buf + (ptr>>2) + 1, (60-ptr) >> 2 );
|
||||
blake32_4way( sc, buf + (ptr>>2), 64 - ptr );
|
||||
sc->T0 = SPH_C32(0xFFFFFE00UL);
|
||||
sc->T1 = SPH_C32(0xFFFFFFFFUL);
|
||||
memset_zero_128( buf, 56>>2 );
|
||||
if (out_size_w32 == 8)
|
||||
buf[52>>2] = _mm_set1_epi32( 0x01000000UL );
|
||||
*(buf+(56>>2)) = mm_bswap_32( _mm_set1_epi32( th ) );
|
||||
*(buf+(60>>2)) = mm_bswap_32( _mm_set1_epi32( tl ) );
|
||||
blake32_4way( sc, buf, 64 );
|
||||
memset_zero_128( buf + vptr + 1, (60-ptr) >> 2 );
|
||||
blake32_4way( ctx, buf + vptr, 64 - ptr );
|
||||
ctx->T0 = 0xFFFFFE00UL;
|
||||
ctx->T1 = 0xFFFFFFFFUL;
|
||||
memset_zero_128( buf, 56>>2 );
|
||||
buf[ 13 ] = _mm_or_si128( buf[ 13 ], _mm_set1_epi32( 0x01000000UL ) );
|
||||
buf[ 14 ] = mm128_bswap_32( _mm_set1_epi32( th ) );
|
||||
buf[ 15 ] = mm128_bswap_32( _mm_set1_epi32( tl ) );
|
||||
blake32_4way( ctx, buf, 64 );
|
||||
}
|
||||
out = (__m128i*)dst;
|
||||
for ( k = 0; k < out_size_w32; k++ )
|
||||
out[k] = mm_bswap_32( sc->H[k] );
|
||||
|
||||
casti_m128i( dst, 0 ) = mm128_bswap_32( casti_m128i( ctx->H, 0 ) );
|
||||
casti_m128i( dst, 1 ) = mm128_bswap_32( casti_m128i( ctx->H, 1 ) );
|
||||
casti_m128i( dst, 2 ) = mm128_bswap_32( casti_m128i( ctx->H, 2 ) );
|
||||
casti_m128i( dst, 3 ) = mm128_bswap_32( casti_m128i( ctx->H, 3 ) );
|
||||
casti_m128i( dst, 4 ) = mm128_bswap_32( casti_m128i( ctx->H, 4 ) );
|
||||
casti_m128i( dst, 5 ) = mm128_bswap_32( casti_m128i( ctx->H, 5 ) );
|
||||
casti_m128i( dst, 6 ) = mm128_bswap_32( casti_m128i( ctx->H, 6 ) );
|
||||
casti_m128i( dst, 7 ) = mm128_bswap_32( casti_m128i( ctx->H, 7 ) );
|
||||
}
|
||||
|
||||
#if defined (__AVX2__)
|
||||
@@ -1217,163 +928,32 @@ blake32_8way_close( blake_8way_small_context *sc, unsigned ub, unsigned n,
|
||||
out[k] = mm256_bswap_32( sc->H[k] );
|
||||
}
|
||||
|
||||
// Blake-512 4 way
|
||||
|
||||
static const sph_u64 salt_zero_big[4] = { 0, 0, 0, 0 };
|
||||
|
||||
static void
|
||||
blake64_4way_init( blake_4way_big_context *sc, const sph_u64 *iv,
|
||||
const sph_u64 *salt )
|
||||
{
|
||||
int i;
|
||||
for ( i = 0; i < 8; i++ )
|
||||
sc->H[i] = _mm256_set1_epi64x( iv[i] );
|
||||
for ( i = 0; i < 4; i++ )
|
||||
sc->S[i] = _mm256_set1_epi64x( salt[i] );
|
||||
sc->T0 = sc->T1 = 0;
|
||||
sc->ptr = 0;
|
||||
}
|
||||
|
||||
static void
|
||||
blake64_4way( blake_4way_big_context *sc, const void *data, size_t len)
|
||||
{
|
||||
__m256i *vdata = (__m256i*)data;
|
||||
__m256i *buf;
|
||||
size_t ptr;
|
||||
DECL_STATE64_4WAY
|
||||
|
||||
const int buf_size = 128; // sizeof/8
|
||||
|
||||
buf = sc->buf;
|
||||
ptr = sc->ptr;
|
||||
if ( len < (buf_size - ptr) )
|
||||
{
|
||||
memcpy_256( buf + (ptr>>3), vdata, len>>3 );
|
||||
ptr += len;
|
||||
sc->ptr = ptr;
|
||||
return;
|
||||
}
|
||||
|
||||
READ_STATE64_4WAY(sc);
|
||||
while ( len > 0 )
|
||||
{
|
||||
size_t clen;
|
||||
|
||||
clen = buf_size - ptr;
|
||||
if ( clen > len )
|
||||
clen = len;
|
||||
memcpy_256( buf + (ptr>>3), vdata, clen>>3 );
|
||||
ptr += clen;
|
||||
vdata = vdata + (clen>>3);
|
||||
len -= clen;
|
||||
if (ptr == buf_size )
|
||||
{
|
||||
if ((T0 = SPH_T64(T0 + 1024)) < 1024)
|
||||
T1 = SPH_T64(T1 + 1);
|
||||
COMPRESS64_4WAY;
|
||||
ptr = 0;
|
||||
}
|
||||
}
|
||||
WRITE_STATE64_4WAY(sc);
|
||||
sc->ptr = ptr;
|
||||
}
|
||||
|
||||
static void
|
||||
blake64_4way_close( blake_4way_big_context *sc,
|
||||
unsigned ub, unsigned n, void *dst, size_t out_size_w64)
|
||||
{
|
||||
// union {
|
||||
__m256i buf[16];
|
||||
// sph_u64 dummy;
|
||||
// } u;
|
||||
size_t ptr, k;
|
||||
unsigned bit_len;
|
||||
uint64_t z, zz;
|
||||
sph_u64 th, tl;
|
||||
__m256i *out;
|
||||
|
||||
ptr = sc->ptr;
|
||||
bit_len = ((unsigned)ptr << 3);
|
||||
z = 0x80 >> n;
|
||||
zz = ((ub & -z) | z) & 0xFF;
|
||||
buf[ptr>>3] = _mm256_set_epi64x( zz, zz, zz, zz );
|
||||
tl = sc->T0 + bit_len;
|
||||
th = sc->T1;
|
||||
if (ptr == 0 )
|
||||
{
|
||||
sc->T0 = SPH_C64(0xFFFFFFFFFFFFFC00ULL);
|
||||
sc->T1 = SPH_C64(0xFFFFFFFFFFFFFFFFULL);
|
||||
}
|
||||
else if ( sc->T0 == 0 )
|
||||
{
|
||||
sc->T0 = SPH_C64(0xFFFFFFFFFFFFFC00ULL) + bit_len;
|
||||
sc->T1 = SPH_T64(sc->T1 - 1);
|
||||
}
|
||||
else
|
||||
{
|
||||
sc->T0 -= 1024 - bit_len;
|
||||
}
|
||||
if ( ptr <= 104 )
|
||||
{
|
||||
memset_zero_256( buf + (ptr>>3) + 1, (104-ptr) >> 3 );
|
||||
if ( out_size_w64 == 8 )
|
||||
buf[(104>>3)] = _mm256_or_si256( buf[(104>>3)],
|
||||
_mm256_set1_epi64x( 0x0100000000000000ULL ) );
|
||||
*(buf+(112>>3)) = mm256_bswap_64(
|
||||
_mm256_set_epi64x( th, th, th, th ) );
|
||||
*(buf+(120>>3)) = mm256_bswap_64(
|
||||
_mm256_set_epi64x( tl, tl, tl, tl ) );
|
||||
|
||||
blake64_4way( sc, buf + (ptr>>3), 128 - ptr );
|
||||
}
|
||||
else
|
||||
{
|
||||
memset_zero_256( buf + (ptr>>3) + 1, (120 - ptr) >> 3 );
|
||||
|
||||
blake64_4way( sc, buf + (ptr>>3), 128 - ptr );
|
||||
sc->T0 = SPH_C64(0xFFFFFFFFFFFFFC00ULL);
|
||||
sc->T1 = SPH_C64(0xFFFFFFFFFFFFFFFFULL);
|
||||
memset_zero_256( buf, 112>>3 );
|
||||
if ( out_size_w64 == 8 )
|
||||
buf[104>>3] = _mm256_set1_epi64x( 0x0100000000000000ULL );
|
||||
*(buf+(112>>3)) = mm256_bswap_64(
|
||||
_mm256_set_epi64x( th, th, th, th ) );
|
||||
*(buf+(120>>3)) = mm256_bswap_64(
|
||||
_mm256_set_epi64x( tl, tl, tl, tl ) );
|
||||
|
||||
blake64_4way( sc, buf, 128 );
|
||||
}
|
||||
out = (__m256i*)dst;
|
||||
for ( k = 0; k < out_size_w64; k++ )
|
||||
out[k] = mm256_bswap_64( sc->H[k] );
|
||||
}
|
||||
|
||||
#endif
|
||||
|
||||
// Blake-256 4 way
|
||||
|
||||
// default 14 rounds, backward copatibility
|
||||
void
|
||||
blake256_4way_init(void *cc)
|
||||
blake256_4way_init(void *ctx)
|
||||
{
|
||||
blake32_4way_init( cc, IV256, salt_zero_4way_small, 14 );
|
||||
blake32_4way_init( ctx, IV256, salt_zero_4way_small, 14 );
|
||||
}
|
||||
|
||||
void
|
||||
blake256_4way(void *cc, const void *data, size_t len)
|
||||
blake256_4way(void *ctx, const void *data, size_t len)
|
||||
{
|
||||
blake32_4way(cc, data, len);
|
||||
blake32_4way(ctx, data, len);
|
||||
}
|
||||
|
||||
void
|
||||
blake256_4way_close(void *cc, void *dst)
|
||||
blake256_4way_close(void *ctx, void *dst)
|
||||
{
|
||||
blake32_4way_close(cc, 0, 0, dst, 8);
|
||||
blake32_4way_close(ctx, 0, 0, dst, 8);
|
||||
}
|
||||
|
||||
#if defined(__AVX2__)
|
||||
|
||||
// Blake-256 8way
|
||||
// Blake-256 8 way
|
||||
|
||||
void
|
||||
blake256_8way_init(void *cc)
|
||||
@@ -1473,38 +1053,8 @@ blake256r8_8way_close(void *cc, void *dst)
|
||||
|
||||
#endif
|
||||
|
||||
// Blake-512 4 way
|
||||
|
||||
#if defined (__AVX2__)
|
||||
|
||||
void
|
||||
blake512_4way_init(void *cc)
|
||||
{
|
||||
blake64_4way_init(cc, IV512, salt_zero_big);
|
||||
}
|
||||
|
||||
void
|
||||
blake512_4way(void *cc, const void *data, size_t len)
|
||||
{
|
||||
blake64_4way(cc, data, len);
|
||||
}
|
||||
|
||||
void
|
||||
blake512_4way_close(void *cc, void *dst)
|
||||
{
|
||||
blake512_4way_addbits_and_close(cc, 0, 0, dst);
|
||||
}
|
||||
|
||||
void
|
||||
blake512_4way_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
|
||||
{
|
||||
blake64_4way_close(cc, ub, n, dst, 8);
|
||||
}
|
||||
|
||||
#endif
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif
|
||||
//#endif
|
||||
322
algo/blake/blake256-hash-4way.c.new
Normal file
322
algo/blake/blake256-hash-4way.c.new
Normal file
@@ -0,0 +1,322 @@
|
||||
// convert blake256 32 bit to use 64 bit with serial vectoring
|
||||
//
|
||||
// cut calls to GS in half
|
||||
//
|
||||
// combine V
|
||||
// v0 = {V0,V1}
|
||||
// v1 = {V2,V3}
|
||||
// v2 = {V4,V5}
|
||||
// v3 = {V6,V7}
|
||||
// v4 = {V8,V9}
|
||||
// v5 = {VA,VB}
|
||||
// v6 = {VC,VD}
|
||||
// v7 = {CE,VF}
|
||||
//
|
||||
// v6x = {VD,VC} swap(VC,VD) swap(v6)
|
||||
// v7x = {VF,VE} swap(VE,VF) swap(v7)
|
||||
//
|
||||
// V0 = v1v0
|
||||
// V1 = v3v2
|
||||
// V2 = v5v4
|
||||
// V3 = v7v6
|
||||
// V4 = v9v8
|
||||
// V5 = vbva
|
||||
// V6 = vdvc
|
||||
// V7 = vfve
|
||||
//
|
||||
// The rotate in ROUND is to effect straddle and unstraddle for the third
|
||||
// and 4th iteration of GS.
|
||||
// It concatenates 2 contiguous 256 bit vectors and extracts the middle
|
||||
// 256 bits. After the transform they must be restored with only the
|
||||
// chosen bits modified in the original 2 vectors.
|
||||
// ror1x128 achieves this by putting the chosen bits in arg1, the "low"
|
||||
// 256 bit vector and saves the untouched bits temporailly in arg0, the
|
||||
// "high" 256 bit vector. Simply reverse the process to restore data back
|
||||
// to original positions.
|
||||
|
||||
// Use standard 4way when AVX2 is not available use x2 mode with AVX2.
|
||||
//
|
||||
// Data is organised the same as 32 bit 4 way, in effect serial vectoring
|
||||
// on top of parallel vectoring. Same data in the same place just taking
|
||||
// two chunks at a time.
|
||||
//
|
||||
// Transparent to user, x2 mode used when AVX2 detected.
|
||||
// Use existing 4way context but revert to scalar types.
|
||||
// Same interleave function (128 bit) or x2 with 256 bit?
|
||||
// User trsnaparency would have to apply to interleave as well.
|
||||
//
|
||||
// Use common 4way update and close
|
||||
|
||||
/*
|
||||
typedef struct {
|
||||
unsigned char buf[64<<2];
|
||||
uint32_t H[8<<2];
|
||||
uint32_t S[4<<2];
|
||||
size_t ptr;
|
||||
uint32_t T0, T1;
|
||||
int rounds; // 14 for blake, 8 for blakecoin & vanilla
|
||||
} blakex2_4way_small_context __attribute__ ((aligned (64)));
|
||||
*/
|
||||
|
||||
static void
|
||||
blake32x2_4way_init( blake_4way_small_context *ctx, const uint32_t *iv,
|
||||
const uint32_t *salt, int rounds )
|
||||
{
|
||||
casti_m128i( ctx->H, 0 ) = _mm_set1_epi32( iv[0] );
|
||||
casti_m128i( ctx->H, 1 ) = _mm_set1_epi32( iv[1] );
|
||||
casti_m128i( ctx->H, 2 ) = _mm_set1_epi32( iv[2] );
|
||||
casti_m128i( ctx->H, 3 ) = _mm_set1_epi32( iv[3] );
|
||||
casti_m128i( ctx->H, 4 ) = _mm_set1_epi32( iv[4] );
|
||||
casti_m128i( ctx->H, 5 ) = _mm_set1_epi32( iv[5] );
|
||||
casti_m128i( ctx->H, 6 ) = _mm_set1_epi32( iv[6] );
|
||||
casti_m128i( ctx->H, 7 ) = _mm_set1_epi32( iv[7] );
|
||||
|
||||
casti_m128i( ctx->S, 0 ) = m128_zero;
|
||||
casti_m128i( ctx->S, 1 ) = m128_zero;
|
||||
casti_m128i( ctx->S, 2 ) = m128_zero;
|
||||
casti_m128i( ctx->S, 3 ) = m128_zero;
|
||||
/*
|
||||
sc->S[0] = _mm_set1_epi32( salt[0] );
|
||||
sc->S[1] = _mm_set1_epi32( salt[1] );
|
||||
sc->S[2] = _mm_set1_epi32( salt[2] );
|
||||
sc->S[3] = _mm_set1_epi32( salt[3] );
|
||||
*/
|
||||
ctx->T0 = ctx->T1 = 0;
|
||||
ctx->ptr = 0;
|
||||
ctx->rounds = rounds;
|
||||
}
|
||||
|
||||
static void
|
||||
blake32x2( blake_4way_small_context *ctx, const void *data, size_t len )
|
||||
{
|
||||
__m128i *buf = (__m256i*)ctx->buf;
|
||||
size_t bptr = ctx->ptr << 2;
|
||||
size_t vptr = ctx->ptr >> 3;
|
||||
size_t blen = len << 2;
|
||||
// unsigned char *buf = ctx->buf;
|
||||
// size_t ptr = ctx->ptr<<4; // repurposed
|
||||
DECL_STATE32x2
|
||||
|
||||
// buf = sc->buf;
|
||||
// ptr = sc->ptr;
|
||||
|
||||
// adjust len for use with ptr, clen, all absolute bytes.
|
||||
// int blen = len<<2;
|
||||
|
||||
if ( blen < (sizeof ctx->buf) - bptr )
|
||||
{
|
||||
memcpy( buf + vptr, data, blen );
|
||||
ptr += blen;
|
||||
ctx->ptr = bptr >> 2;;
|
||||
return;
|
||||
}
|
||||
|
||||
READ_STATE32( ctx );
|
||||
while ( blen > 0 )
|
||||
{
|
||||
size_t clen;
|
||||
|
||||
clen = ( sizeof sc->buf ) - ptr;
|
||||
if ( clen > blen )
|
||||
clen = blen;
|
||||
memcpy( buf + vptr, data, clen );
|
||||
bptr += clen;
|
||||
vptr = bptr >> 5;
|
||||
data = (const unsigned char *)data + clen;
|
||||
blen -= clen;
|
||||
if ( bptr == sizeof ctx->buf )
|
||||
{
|
||||
if ( ( T0 = T0 + 512 ) < 512 ) // not needed, will never rollover
|
||||
T1 += 1;
|
||||
COMPRESS32x2_4WAY( ctx->rounds );
|
||||
ptr = 0;
|
||||
}
|
||||
}
|
||||
WRITE_STATE32x2( ctx );
|
||||
ctx->ptr = bptr >> 2;
|
||||
}
|
||||
|
||||
static void
|
||||
blake32x2_4way_close( blake_4way_small_context *ctx, void *dst )
|
||||
{
|
||||
__m256i buf[8] __attribute__ ((aligned (64)));
|
||||
size_t ptr = ctx->ptr;
|
||||
size_t vptr = ctx->ptr>>2;
|
||||
unsigned bit_len = ( (unsigned)ptr << 3 ); // one lane
|
||||
uint32_t th = ctx->T1;
|
||||
uint32_t tl = ctx->T0 + bit_len;
|
||||
|
||||
if ( ptr == 0 )
|
||||
{
|
||||
ctx->T0 = 0xFFFFFE00UL;
|
||||
ctx->T1 = 0xFFFFFFFFUL;
|
||||
}
|
||||
else if ( ctx->T0 == 0 )
|
||||
{
|
||||
ctx->T0 = 0xFFFFFE00UL + bit_len;
|
||||
ctx->T1 -= 1;
|
||||
}
|
||||
else
|
||||
ctx->T0 -= 512 - bit_len;
|
||||
|
||||
// memset doesn't do ints
|
||||
buf[ vptr ] = _mm256_set_epi32( 0,0,0,0, 0x80, 0x80, 0x80, 0x80 );
|
||||
|
||||
if ( vptr < 5 )
|
||||
{
|
||||
memset_zero_256( buf + vptr + 1, 6 - vptr );
|
||||
buf[ 6 ] = _mm256_or_si256( vbuf[ 6 ], _mm256_set_epi32(
|
||||
0x01000000UL,0x01000000UL,0x01000000UL,0x01000000UL, 0,0,0,0 ) );
|
||||
buf[ 7 ] = mm256_bswap_32( _mm256_set_epi32( tl,tl,tl,tl,
|
||||
th,th,th,th ) );
|
||||
blake32x2_4way( ctx, buf + vptr, 64 - ptr );
|
||||
}
|
||||
else
|
||||
{
|
||||
memset_zero_256( vbuf + vptr + 1, 7 - vptr );
|
||||
blake32x2_4way( ctx, vbuf + ptr, 64 - ptr );
|
||||
ctx->T0 = 0xFFFFFE00UL;
|
||||
ctx->T1 = 0xFFFFFFFFUL;
|
||||
buf[ 6 ] = mm256_zero;
|
||||
buf[ 6 ] = _mm256_set_epi32( 0,0,0,0,
|
||||
0x01000000UL,0x01000000UL,0x01000000UL,0x01000000UL );
|
||||
buf[ 7 ] = mm256_bswap_32( _mm256_set_epi32( tl, tl, tl, tl,
|
||||
th, th, th, th );
|
||||
blake32x2_4way( ctx, buf, 64 );
|
||||
}
|
||||
|
||||
casti_m256i( dst, 0 ) = mm256_bswap_32( casti_m256i( ctx->H, 0 ) );
|
||||
casti_m256i( dst, 1 ) = mm256_bswap_32( casti_m256i( ctx->H, 1 ) );
|
||||
casti_m256i( dst, 2 ) = mm256_bswap_32( casti_m256i( ctx->H, 2 ) );
|
||||
casti_m256i( dst, 3 ) = mm256_bswap_32( casti_m256i( ctx->H, 3 ) );
|
||||
}
|
||||
|
||||
|
||||
|
||||
|
||||
#define DECL_STATE32x2_4WAY \
|
||||
__m256i H0, H1, H2, H3; \
|
||||
__m256i S0, S1; \
|
||||
uint32_t T0, T1;
|
||||
|
||||
#define READ_STATE32x2_4WAY(state) do \
|
||||
{ \
|
||||
H0 = casti_m256i( state->H, 0 ); \
|
||||
H1 = casti_m256i( state->H, 1 ); \
|
||||
H2 = casti_m256i( state->H, 2 ); \
|
||||
H3 = casti_m256i( state->H, 3 ); \
|
||||
S0 = casti_m256i( state->S, 0 ); \
|
||||
S1 = casti_m256i( state->S, 1 ); \
|
||||
T0 = state->T0; \
|
||||
T1 = state->T1; \
|
||||
|
||||
#define WRITE_STATE32x2_4WAY(state) do { \
|
||||
casti_m256i( state->H, 0 ) = H0; \
|
||||
casti_m256i( state->H, 1 ) = H1; \
|
||||
casti_m256i( state->H, 2 ) = H2; \
|
||||
casti_m256i( state->H, 3 ) = H3; \
|
||||
casti_m256i( state->S, 0 ) = S0; \
|
||||
casti_m256i( state->S, 1 ) = S1; \
|
||||
state->T0 = T0; \
|
||||
state->T1 = T1; \
|
||||
} while (0)
|
||||
|
||||
|
||||
#define GSx2_4WAY( m0m2, m1m3, c0c2, c1c3, a, b, c, d ) do \
|
||||
{ \
|
||||
a = _mm256_add_epi32( _mm256_add_epi32( _mm256_xor_si256( \
|
||||
_mm256_set_epi32( c1,c3, c1,c3, c1,c3, c1,c3 ), \
|
||||
_mm256_set_epi32( m0,m2, m0,m2, m0,m2, m0,m2 ) ), b ), a ); \
|
||||
d = mm256_ror_32( _mm_xor_si128( d, a ), 16 ); \
|
||||
c = _mm256_add_epi32( c, d ); \
|
||||
b = mm256_ror_32( _mm256_xor_si256( b, c ), 12 ); \
|
||||
a = _mm256_add_epi32( _mm256_add_epi32( _mm256_xor_si256( \
|
||||
_mm256_set_epi32( c0,c2, c0,c2, c0,c2, c0,c2 ), \
|
||||
_mm256_set_epi32( m1,m3, m1,m3, m1,m3, m1,m3 ) ), b ), a ); \
|
||||
d = mm256_ror_32( _mm256_xor_si256( d, a ), 8 ); \
|
||||
c = _mm256_add_epi32( c, d ); \
|
||||
b = mm256_ror_32( _mm256_xor_si256( b, c ), 7 ); \
|
||||
} while (0)
|
||||
|
||||
#define ROUND_Sx2_4WAY(r) do \
|
||||
{ \
|
||||
GS2_4WAY( Mx(r, 0), Mx(r, 1), Mx(r, 2), Mx(r, 3), \
|
||||
CSx(r, 0), CSx(r, 1), CSx(r, 2), CSx(r, 3), V0, V2, V4, V6 ); \
|
||||
GS2_4WAY( Mx(r, 4), Mx(r, 5), Mx(r, 6), Mx(r, 7), \
|
||||
CSx(r, 4), CSx(r, 5), CSx(r, 6), CSx(r, 7), V1, V3, V5, V7 ); \
|
||||
mm256_ror1x128_512( V3, V2 ); \
|
||||
mm256_ror1x128_512( V6, V7 ); \
|
||||
GS2_4WAY( Mx(r, 8), Mx(r, 9), Mx(r, A), Mx(r, B), \
|
||||
CSx(r, 8), CSx(r, 9), CSx(r, A), CSx(r, B), V0, V2, V5, V7 ); \
|
||||
GS2_4WAY( Mx(r, C), Mx(r, D), Mx(r, C), Mx(r, D), \
|
||||
CSx(r, C), CSx(r, D), CSx(r, C), CSx(r, D), V1, V3, V4, V6 ); \
|
||||
mm256_rol1x128_512( V2, V3 ); \
|
||||
mm256_rol1x128_512( V7, V6 );
|
||||
|
||||
#define COMPRESS32x2_4WAY( rounds ) do \
|
||||
{ \
|
||||
__m256i M0, M1, M2, M3, M4, M5, M6, M7; \
|
||||
__m256i V0, V1, V2, V3, V4, V5, V6, V7; \
|
||||
unsigned r; \
|
||||
V0 = H0; \
|
||||
V1 = H1; \
|
||||
V2 = H2; \
|
||||
V3 = H3; \
|
||||
V4 = _mm256_xor_si256( S0, _mm256_set_epi32( CS1, CS1, CS1, CS1, \
|
||||
CS0, CS0, CS0, CS0 ) ); \
|
||||
V5 = _mm256_xor_si256( S1, _mm256_set_epi32( CS3, CS3, CS3, CS3, \
|
||||
CS2, CS2, CS2, CS2 ) ); \
|
||||
V6 = _mm256_xor_si256( _mm256_set1_epi32( T0 ), \
|
||||
_mm256_set_epi32( CS5, CS5, CS5, CS5, \
|
||||
CS4, CS4, CS4, CS4 ) ); \
|
||||
V7 = _mm256_xor_si256( _mm256_set1_epi32( T1 ), \
|
||||
_mm256_set_epi32( CS7, CS7, CS7, CS7, \
|
||||
CS6, CS6, CS6, CS6 ) ); \
|
||||
M0 = mm256_bswap_32( buf[ 0] ); \
|
||||
M1 = mm256_bswap_32( buf[ 1] ); \
|
||||
M2 = mm256_bswap_32( buf[ 2] ); \
|
||||
M3 = mm256_bswap_32( buf[ 3] ); \
|
||||
M4 = mm256_bswap_32( buf[ 4] ); \
|
||||
M5 = mm256_bswap_32( buf[ 5] ); \
|
||||
M6 = mm256_bswap_32( buf[ 6] ); \
|
||||
M7 = mm256_bswap_32( buf[ 7] ); \
|
||||
ROUND_Sx2_4WAY(0); \
|
||||
ROUND_Sx2_4WAY(1); \
|
||||
ROUND_Sx2_4WAY(2); \
|
||||
ROUND_Sx2_4WAY(3); \
|
||||
ROUND_Sx2_4WAY(4); \
|
||||
ROUND_Sx2_4WAY(5); \
|
||||
ROUND_Sx2_4WAY(6); \
|
||||
ROUND_Sx2_4WAY(7); \
|
||||
if (rounds == 14) \
|
||||
{ \
|
||||
ROUND_Sx2_4WAY(8); \
|
||||
ROUND_Sx2_4WAY(9); \
|
||||
ROUND_Sx2_4WAY(0); \
|
||||
ROUND_Sx2_4WAY(1); \
|
||||
ROUND_Sx2_4WAY(2); \
|
||||
ROUND_Sx2_4WAY(3); \
|
||||
} \
|
||||
H0 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( V8, V0 ), S0 ), H0 ); \
|
||||
H1 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( V9, V1 ), S1 ), H1 ); \
|
||||
H2 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( VA, V2 ), S2 ), H2 ); \
|
||||
H3 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( VB, V3 ), S3 ), H3 ); \
|
||||
} while (0)
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
@@ -85,7 +85,8 @@ void blake2s_4way_hash( void *output, const void *input )
|
||||
blake2s_4way_update( &ctx, input + (64<<2), 16 );
|
||||
blake2s_4way_final( &ctx, vhash, BLAKE2S_OUTBYTES );
|
||||
|
||||
mm_deinterleave_4x32( output, output+32, output+64, output+96, vhash, 256 );
|
||||
mm128_deinterleave_4x32( output, output+32, output+64, output+96,
|
||||
vhash, 256 );
|
||||
}
|
||||
|
||||
int scanhash_blake2s_4way( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
@@ -104,7 +105,7 @@ int scanhash_blake2s_4way( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
uint32_t *noncep = vdata + 76; // 19*4
|
||||
|
||||
swab32_array( edata, pdata, 20 );
|
||||
mm_interleave_4x32( vdata, edata, edata, edata, edata, 640 );
|
||||
mm128_interleave_4x32( vdata, edata, edata, edata, edata, 640 );
|
||||
blake2s_4way_init( &blake2s_4w_ctx, BLAKE2S_OUTBYTES );
|
||||
blake2s_4way_update( &blake2s_4w_ctx, vdata, 64 );
|
||||
|
||||
|
||||
@@ -20,7 +20,7 @@ bool register_blake2s_algo( algo_gate_t* gate )
|
||||
gate->hash = (void*)&blake2s_hash;
|
||||
#endif
|
||||
gate->get_max64 = (void*)&blake2s_get_max64;
|
||||
gate->optimizations = AVX_OPT | AVX2_OPT;
|
||||
gate->optimizations = SSE42_OPT | AVX2_OPT;
|
||||
return true;
|
||||
};
|
||||
|
||||
|
||||
@@ -4,7 +4,7 @@
|
||||
#include <stdint.h>
|
||||
#include "algo-gate-api.h"
|
||||
|
||||
#if defined(__AVX__)
|
||||
#if defined(__SSE4_2__)
|
||||
#define BLAKE2S_4WAY
|
||||
#endif
|
||||
#if defined(__AVX2__)
|
||||
|
||||
@@ -17,7 +17,7 @@
|
||||
#include <string.h>
|
||||
#include <stdio.h>
|
||||
|
||||
#if defined(__AVX__)
|
||||
#if defined(__SSE4_2__)
|
||||
|
||||
static const uint32_t blake2s_IV[8] =
|
||||
{
|
||||
@@ -92,13 +92,13 @@ int blake2s_4way_compress( blake2s_4way_state *S, const __m128i* block )
|
||||
#define G4W(r,i,a,b,c,d) \
|
||||
do { \
|
||||
a = _mm_add_epi32( _mm_add_epi32( a, b ), m[ blake2s_sigma[r][2*i+0] ] ); \
|
||||
d = mm_rotr_32( _mm_xor_si128( d, a ), 16 ); \
|
||||
d = mm128_ror_32( _mm_xor_si128( d, a ), 16 ); \
|
||||
c = _mm_add_epi32( c, d ); \
|
||||
b = mm_rotr_32( _mm_xor_si128( b, c ), 12 ); \
|
||||
b = mm128_ror_32( _mm_xor_si128( b, c ), 12 ); \
|
||||
a = _mm_add_epi32( _mm_add_epi32( a, b ), m[ blake2s_sigma[r][2*i+1] ] ); \
|
||||
d = mm_rotr_32( _mm_xor_si128( d, a ), 8 ); \
|
||||
d = mm128_ror_32( _mm_xor_si128( d, a ), 8 ); \
|
||||
c = _mm_add_epi32( c, d ); \
|
||||
b = mm_rotr_32( _mm_xor_si128( b, c ), 7 ); \
|
||||
b = mm128_ror_32( _mm_xor_si128( b, c ), 7 ); \
|
||||
} while(0)
|
||||
|
||||
#define ROUND4W(r) \
|
||||
@@ -210,14 +210,14 @@ int blake2s_8way_compress( blake2s_8way_state *S, const __m256i *block )
|
||||
do { \
|
||||
a = _mm256_add_epi32( _mm256_add_epi32( a, b ), \
|
||||
m[ blake2s_sigma[r][2*i+0] ] ); \
|
||||
d = mm256_rotr_32( _mm256_xor_si256( d, a ), 16 ); \
|
||||
d = mm256_ror_32( _mm256_xor_si256( d, a ), 16 ); \
|
||||
c = _mm256_add_epi32( c, d ); \
|
||||
b = mm256_rotr_32( _mm256_xor_si256( b, c ), 12 ); \
|
||||
b = mm256_ror_32( _mm256_xor_si256( b, c ), 12 ); \
|
||||
a = _mm256_add_epi32( _mm256_add_epi32( a, b ), \
|
||||
m[ blake2s_sigma[r][2*i+1] ] ); \
|
||||
d = mm256_rotr_32( _mm256_xor_si256( d, a ), 8 ); \
|
||||
d = mm256_ror_32( _mm256_xor_si256( d, a ), 8 ); \
|
||||
c = _mm256_add_epi32( c, d ); \
|
||||
b = mm256_rotr_32( _mm256_xor_si256( b, c ), 7 ); \
|
||||
b = mm256_ror_32( _mm256_xor_si256( b, c ), 7 ); \
|
||||
} while(0)
|
||||
|
||||
#define ROUND8W(r) \
|
||||
@@ -359,4 +359,4 @@ int blake2s( uint8_t *out, const void *in, const void *key, const uint8_t outlen
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif // __AVX__
|
||||
#endif // __SSE4_2__
|
||||
|
||||
@@ -14,9 +14,9 @@
|
||||
#ifndef __BLAKE2S_HASH_4WAY_H__
|
||||
#define __BLAKE2S_HASH_4WAY_H__ 1
|
||||
|
||||
#if defined(__AVX__)
|
||||
#if defined(__SSE4_2__)
|
||||
|
||||
#include "avxdefs.h"
|
||||
#include "simd-utils.h"
|
||||
|
||||
#include <stddef.h>
|
||||
#include <stdint.h>
|
||||
@@ -107,6 +107,6 @@ int blake2s_8way_final( blake2s_8way_state *S, void *out, uint8_t outlen );
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif // __AVX__
|
||||
#endif // __SSE4_2__
|
||||
|
||||
#endif
|
||||
|
||||
701
algo/blake/blake512-hash-4way.c
Normal file
701
algo/blake/blake512-hash-4way.c
Normal file
@@ -0,0 +1,701 @@
|
||||
/* $Id: blake.c 252 2011-06-07 17:55:14Z tp $ */
|
||||
/*
|
||||
* BLAKE implementation.
|
||||
*
|
||||
* ==========================(LICENSE BEGIN)============================
|
||||
*
|
||||
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
|
||||
*
|
||||
* Permission is hereby granted, free of charge, to any person obtaining
|
||||
* a copy of this software and associated documentation files (the
|
||||
* "Software"), to deal in the Software without restriction, including
|
||||
* without limitation the rights to use, copy, modify, merge, publish,
|
||||
* distribute, sublicense, and/or sell copies of the Software, and to
|
||||
* permit persons to whom the Software is furnished to do so, subject to
|
||||
* the following conditions:
|
||||
*
|
||||
* The above copyright notice and this permission notice shall be
|
||||
* included in all copies or substantial portions of the Software.
|
||||
*
|
||||
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
||||
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
||||
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
|
||||
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
|
||||
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
|
||||
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
|
||||
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
||||
*
|
||||
* ===========================(LICENSE END)=============================
|
||||
*
|
||||
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
|
||||
*/
|
||||
|
||||
#if defined (__AVX2__)
|
||||
|
||||
#include <stddef.h>
|
||||
#include <string.h>
|
||||
#include <limits.h>
|
||||
|
||||
#include "blake-hash-4way.h"
|
||||
|
||||
#ifdef __cplusplus
|
||||
extern "C"{
|
||||
#endif
|
||||
|
||||
#if SPH_SMALL_FOOTPRINT && !defined SPH_SMALL_FOOTPRINT_BLAKE
|
||||
#define SPH_SMALL_FOOTPRINT_BLAKE 1
|
||||
#endif
|
||||
|
||||
#if SPH_64 && (SPH_SMALL_FOOTPRINT_BLAKE || !SPH_64_TRUE)
|
||||
#define SPH_COMPACT_BLAKE_64 1
|
||||
#endif
|
||||
|
||||
#ifdef _MSC_VER
|
||||
#pragma warning (disable: 4146)
|
||||
#endif
|
||||
|
||||
|
||||
// Blake-512
|
||||
|
||||
static const sph_u64 IV512[8] = {
|
||||
SPH_C64(0x6A09E667F3BCC908), SPH_C64(0xBB67AE8584CAA73B),
|
||||
SPH_C64(0x3C6EF372FE94F82B), SPH_C64(0xA54FF53A5F1D36F1),
|
||||
SPH_C64(0x510E527FADE682D1), SPH_C64(0x9B05688C2B3E6C1F),
|
||||
SPH_C64(0x1F83D9ABFB41BD6B), SPH_C64(0x5BE0CD19137E2179)
|
||||
};
|
||||
|
||||
|
||||
#if SPH_COMPACT_BLAKE_32 || SPH_COMPACT_BLAKE_64
|
||||
|
||||
// Blake-256 4 & 8 way, Blake-512 4 way
|
||||
|
||||
static const unsigned sigma[16][16] = {
|
||||
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 },
|
||||
{ 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 },
|
||||
{ 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 },
|
||||
{ 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 },
|
||||
{ 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13 },
|
||||
{ 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9 },
|
||||
{ 12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11 },
|
||||
{ 13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10 },
|
||||
{ 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5 },
|
||||
{ 10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13, 0 },
|
||||
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 },
|
||||
{ 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 },
|
||||
{ 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 },
|
||||
{ 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 },
|
||||
{ 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13 },
|
||||
{ 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9 }
|
||||
};
|
||||
|
||||
#endif
|
||||
|
||||
#define Z00 0
|
||||
#define Z01 1
|
||||
#define Z02 2
|
||||
#define Z03 3
|
||||
#define Z04 4
|
||||
#define Z05 5
|
||||
#define Z06 6
|
||||
#define Z07 7
|
||||
#define Z08 8
|
||||
#define Z09 9
|
||||
#define Z0A A
|
||||
#define Z0B B
|
||||
#define Z0C C
|
||||
#define Z0D D
|
||||
#define Z0E E
|
||||
#define Z0F F
|
||||
|
||||
#define Z10 E
|
||||
#define Z11 A
|
||||
#define Z12 4
|
||||
#define Z13 8
|
||||
#define Z14 9
|
||||
#define Z15 F
|
||||
#define Z16 D
|
||||
#define Z17 6
|
||||
#define Z18 1
|
||||
#define Z19 C
|
||||
#define Z1A 0
|
||||
#define Z1B 2
|
||||
#define Z1C B
|
||||
#define Z1D 7
|
||||
#define Z1E 5
|
||||
#define Z1F 3
|
||||
|
||||
#define Z20 B
|
||||
#define Z21 8
|
||||
#define Z22 C
|
||||
#define Z23 0
|
||||
#define Z24 5
|
||||
#define Z25 2
|
||||
#define Z26 F
|
||||
#define Z27 D
|
||||
#define Z28 A
|
||||
#define Z29 E
|
||||
#define Z2A 3
|
||||
#define Z2B 6
|
||||
#define Z2C 7
|
||||
#define Z2D 1
|
||||
#define Z2E 9
|
||||
#define Z2F 4
|
||||
|
||||
#define Z30 7
|
||||
#define Z31 9
|
||||
#define Z32 3
|
||||
#define Z33 1
|
||||
#define Z34 D
|
||||
#define Z35 C
|
||||
#define Z36 B
|
||||
#define Z37 E
|
||||
#define Z38 2
|
||||
#define Z39 6
|
||||
#define Z3A 5
|
||||
#define Z3B A
|
||||
#define Z3C 4
|
||||
#define Z3D 0
|
||||
#define Z3E F
|
||||
#define Z3F 8
|
||||
|
||||
#define Z40 9
|
||||
#define Z41 0
|
||||
#define Z42 5
|
||||
#define Z43 7
|
||||
#define Z44 2
|
||||
#define Z45 4
|
||||
#define Z46 A
|
||||
#define Z47 F
|
||||
#define Z48 E
|
||||
#define Z49 1
|
||||
#define Z4A B
|
||||
#define Z4B C
|
||||
#define Z4C 6
|
||||
#define Z4D 8
|
||||
#define Z4E 3
|
||||
#define Z4F D
|
||||
|
||||
#define Z50 2
|
||||
#define Z51 C
|
||||
#define Z52 6
|
||||
#define Z53 A
|
||||
#define Z54 0
|
||||
#define Z55 B
|
||||
#define Z56 8
|
||||
#define Z57 3
|
||||
#define Z58 4
|
||||
#define Z59 D
|
||||
#define Z5A 7
|
||||
#define Z5B 5
|
||||
#define Z5C F
|
||||
#define Z5D E
|
||||
#define Z5E 1
|
||||
#define Z5F 9
|
||||
|
||||
#define Z60 C
|
||||
#define Z61 5
|
||||
#define Z62 1
|
||||
#define Z63 F
|
||||
#define Z64 E
|
||||
#define Z65 D
|
||||
#define Z66 4
|
||||
#define Z67 A
|
||||
#define Z68 0
|
||||
#define Z69 7
|
||||
#define Z6A 6
|
||||
#define Z6B 3
|
||||
#define Z6C 9
|
||||
#define Z6D 2
|
||||
#define Z6E 8
|
||||
#define Z6F B
|
||||
|
||||
#define Z70 D
|
||||
#define Z71 B
|
||||
#define Z72 7
|
||||
#define Z73 E
|
||||
#define Z74 C
|
||||
#define Z75 1
|
||||
#define Z76 3
|
||||
#define Z77 9
|
||||
#define Z78 5
|
||||
#define Z79 0
|
||||
#define Z7A F
|
||||
#define Z7B 4
|
||||
#define Z7C 8
|
||||
#define Z7D 6
|
||||
#define Z7E 2
|
||||
#define Z7F A
|
||||
|
||||
#define Z80 6
|
||||
#define Z81 F
|
||||
#define Z82 E
|
||||
#define Z83 9
|
||||
#define Z84 B
|
||||
#define Z85 3
|
||||
#define Z86 0
|
||||
#define Z87 8
|
||||
#define Z88 C
|
||||
#define Z89 2
|
||||
#define Z8A D
|
||||
#define Z8B 7
|
||||
#define Z8C 1
|
||||
#define Z8D 4
|
||||
#define Z8E A
|
||||
#define Z8F 5
|
||||
|
||||
#define Z90 A
|
||||
#define Z91 2
|
||||
#define Z92 8
|
||||
#define Z93 4
|
||||
#define Z94 7
|
||||
#define Z95 6
|
||||
#define Z96 1
|
||||
#define Z97 5
|
||||
#define Z98 F
|
||||
#define Z99 B
|
||||
#define Z9A 9
|
||||
#define Z9B E
|
||||
#define Z9C 3
|
||||
#define Z9D C
|
||||
#define Z9E D
|
||||
#define Z9F 0
|
||||
|
||||
#define Mx(r, i) Mx_(Z ## r ## i)
|
||||
#define Mx_(n) Mx__(n)
|
||||
#define Mx__(n) M ## n
|
||||
|
||||
// Blake-512 4 way
|
||||
|
||||
#define CBx(r, i) CBx_(Z ## r ## i)
|
||||
#define CBx_(n) CBx__(n)
|
||||
#define CBx__(n) CB ## n
|
||||
|
||||
#define CB0 SPH_C64(0x243F6A8885A308D3)
|
||||
#define CB1 SPH_C64(0x13198A2E03707344)
|
||||
#define CB2 SPH_C64(0xA4093822299F31D0)
|
||||
#define CB3 SPH_C64(0x082EFA98EC4E6C89)
|
||||
#define CB4 SPH_C64(0x452821E638D01377)
|
||||
#define CB5 SPH_C64(0xBE5466CF34E90C6C)
|
||||
#define CB6 SPH_C64(0xC0AC29B7C97C50DD)
|
||||
#define CB7 SPH_C64(0x3F84D5B5B5470917)
|
||||
#define CB8 SPH_C64(0x9216D5D98979FB1B)
|
||||
#define CB9 SPH_C64(0xD1310BA698DFB5AC)
|
||||
#define CBA SPH_C64(0x2FFD72DBD01ADFB7)
|
||||
#define CBB SPH_C64(0xB8E1AFED6A267E96)
|
||||
#define CBC SPH_C64(0xBA7C9045F12C7F99)
|
||||
#define CBD SPH_C64(0x24A19947B3916CF7)
|
||||
#define CBE SPH_C64(0x0801F2E2858EFC16)
|
||||
#define CBF SPH_C64(0x636920D871574E69)
|
||||
|
||||
#if SPH_COMPACT_BLAKE_64
|
||||
// not used
|
||||
static const sph_u64 CB[16] = {
|
||||
SPH_C64(0x243F6A8885A308D3), SPH_C64(0x13198A2E03707344),
|
||||
SPH_C64(0xA4093822299F31D0), SPH_C64(0x082EFA98EC4E6C89),
|
||||
SPH_C64(0x452821E638D01377), SPH_C64(0xBE5466CF34E90C6C),
|
||||
SPH_C64(0xC0AC29B7C97C50DD), SPH_C64(0x3F84D5B5B5470917),
|
||||
SPH_C64(0x9216D5D98979FB1B), SPH_C64(0xD1310BA698DFB5AC),
|
||||
SPH_C64(0x2FFD72DBD01ADFB7), SPH_C64(0xB8E1AFED6A267E96),
|
||||
SPH_C64(0xBA7C9045F12C7F99), SPH_C64(0x24A19947B3916CF7),
|
||||
SPH_C64(0x0801F2E2858EFC16), SPH_C64(0x636920D871574E69)
|
||||
};
|
||||
|
||||
#endif
|
||||
|
||||
|
||||
// Blake-512 4 way
|
||||
|
||||
#define GB_4WAY(m0, m1, c0, c1, a, b, c, d) do { \
|
||||
a = _mm256_add_epi64( _mm256_add_epi64( _mm256_xor_si256( \
|
||||
_mm256_set_epi64x( c1, c1, c1, c1 ), m0 ), b ), a ); \
|
||||
d = mm256_ror_64( _mm256_xor_si256( d, a ), 32 ); \
|
||||
c = _mm256_add_epi64( c, d ); \
|
||||
b = mm256_ror_64( _mm256_xor_si256( b, c ), 25 ); \
|
||||
a = _mm256_add_epi64( _mm256_add_epi64( _mm256_xor_si256( \
|
||||
_mm256_set_epi64x( c0, c0, c0, c0 ), m1 ), b ), a ); \
|
||||
d = mm256_ror_64( _mm256_xor_si256( d, a ), 16 ); \
|
||||
c = _mm256_add_epi64( c, d ); \
|
||||
b = mm256_ror_64( _mm256_xor_si256( b, c ), 11 ); \
|
||||
} while (0)
|
||||
|
||||
#if SPH_COMPACT_BLAKE_64
|
||||
// not used
|
||||
#define ROUND_B_4WAY(r) do { \
|
||||
GB_4WAY(M[sigma[r][0x0]], M[sigma[r][0x1]], \
|
||||
CB[sigma[r][0x0]], CB[sigma[r][0x1]], V0, V4, V8, VC); \
|
||||
GB_4WAY(M[sigma[r][0x2]], M[sigma[r][0x3]], \
|
||||
CB[sigma[r][0x2]], CB[sigma[r][0x3]], V1, V5, V9, VD); \
|
||||
GB_4WAY(M[sigma[r][0x4]], M[sigma[r][0x5]], \
|
||||
CB[sigma[r][0x4]], CB[sigma[r][0x5]], V2, V6, VA, VE); \
|
||||
GB_4WAY(M[sigma[r][0x6]], M[sigma[r][0x7]], \
|
||||
CB[sigma[r][0x6]], CB[sigma[r][0x7]], V3, V7, VB, VF); \
|
||||
GB_4WAY(M[sigma[r][0x8]], M[sigma[r][0x9]], \
|
||||
CB[sigma[r][0x8]], CB[sigma[r][0x9]], V0, V5, VA, VF); \
|
||||
GB_4WAY(M[sigma[r][0xA]], M[sigma[r][0xB]], \
|
||||
CB[sigma[r][0xA]], CB[sigma[r][0xB]], V1, V6, VB, VC); \
|
||||
GB_4WAY(M[sigma[r][0xC]], M[sigma[r][0xD]], \
|
||||
CB[sigma[r][0xC]], CB[sigma[r][0xD]], V2, V7, V8, VD); \
|
||||
GB_4WAY(M[sigma[r][0xE]], M[sigma[r][0xF]], \
|
||||
CB[sigma[r][0xE]], CB[sigma[r][0xF]], V3, V4, V9, VE); \
|
||||
} while (0)
|
||||
|
||||
#else
|
||||
//current_impl
|
||||
#define ROUND_B_4WAY(r) do { \
|
||||
GB_4WAY(Mx(r, 0), Mx(r, 1), CBx(r, 0), CBx(r, 1), V0, V4, V8, VC); \
|
||||
GB_4WAY(Mx(r, 2), Mx(r, 3), CBx(r, 2), CBx(r, 3), V1, V5, V9, VD); \
|
||||
GB_4WAY(Mx(r, 4), Mx(r, 5), CBx(r, 4), CBx(r, 5), V2, V6, VA, VE); \
|
||||
GB_4WAY(Mx(r, 6), Mx(r, 7), CBx(r, 6), CBx(r, 7), V3, V7, VB, VF); \
|
||||
GB_4WAY(Mx(r, 8), Mx(r, 9), CBx(r, 8), CBx(r, 9), V0, V5, VA, VF); \
|
||||
GB_4WAY(Mx(r, A), Mx(r, B), CBx(r, A), CBx(r, B), V1, V6, VB, VC); \
|
||||
GB_4WAY(Mx(r, C), Mx(r, D), CBx(r, C), CBx(r, D), V2, V7, V8, VD); \
|
||||
GB_4WAY(Mx(r, E), Mx(r, F), CBx(r, E), CBx(r, F), V3, V4, V9, VE); \
|
||||
} while (0)
|
||||
|
||||
#endif
|
||||
|
||||
|
||||
// Blake-512 4 way
|
||||
|
||||
#define DECL_STATE64_4WAY \
|
||||
__m256i H0, H1, H2, H3, H4, H5, H6, H7; \
|
||||
__m256i S0, S1, S2, S3; \
|
||||
sph_u64 T0, T1;
|
||||
|
||||
#define READ_STATE64_4WAY(state) do { \
|
||||
H0 = (state)->H[0]; \
|
||||
H1 = (state)->H[1]; \
|
||||
H2 = (state)->H[2]; \
|
||||
H3 = (state)->H[3]; \
|
||||
H4 = (state)->H[4]; \
|
||||
H5 = (state)->H[5]; \
|
||||
H6 = (state)->H[6]; \
|
||||
H7 = (state)->H[7]; \
|
||||
S0 = (state)->S[0]; \
|
||||
S1 = (state)->S[1]; \
|
||||
S2 = (state)->S[2]; \
|
||||
S3 = (state)->S[3]; \
|
||||
T0 = (state)->T0; \
|
||||
T1 = (state)->T1; \
|
||||
} while (0)
|
||||
|
||||
#define WRITE_STATE64_4WAY(state) do { \
|
||||
(state)->H[0] = H0; \
|
||||
(state)->H[1] = H1; \
|
||||
(state)->H[2] = H2; \
|
||||
(state)->H[3] = H3; \
|
||||
(state)->H[4] = H4; \
|
||||
(state)->H[5] = H5; \
|
||||
(state)->H[6] = H6; \
|
||||
(state)->H[7] = H7; \
|
||||
(state)->S[0] = S0; \
|
||||
(state)->S[1] = S1; \
|
||||
(state)->S[2] = S2; \
|
||||
(state)->S[3] = S3; \
|
||||
(state)->T0 = T0; \
|
||||
(state)->T1 = T1; \
|
||||
} while (0)
|
||||
|
||||
#if SPH_COMPACT_BLAKE_64
|
||||
|
||||
// not used
|
||||
#define COMPRESS64_4WAY do { \
|
||||
__m256i M[16]; \
|
||||
__m256i V0, V1, V2, V3, V4, V5, V6, V7; \
|
||||
__m256i V8, V9, VA, VB, VC, VD, VE, VF; \
|
||||
unsigned r; \
|
||||
V0 = H0; \
|
||||
V1 = H1; \
|
||||
V2 = H2; \
|
||||
V3 = H3; \
|
||||
V4 = H4; \
|
||||
V5 = H5; \
|
||||
V6 = H6; \
|
||||
V7 = H7; \
|
||||
V8 = _mm256_xor_si256( S0, _mm256_set_epi64x( CB0, CB0, CB0, CB0 ) ); \
|
||||
V9 = _mm256_xor_si256( S1, _mm256_set_epi64x( CB1, CB1, CB1, CB1 ) ); \
|
||||
VA = _mm256_xor_si256( S2, _mm256_set_epi64x( CB2, CB2, CB2, CB2 ) ); \
|
||||
VB = _mm256_xor_si256( S3, _mm256_set_epi64x( CB3, CB3, CB3, CB3 ) ); \
|
||||
VC = _mm256_xor_si256( _mm256_set_epi64x( T0, T0, T0, T0 ), \
|
||||
_mm256_set_epi64x( CB4, CB4, CB4, CB4 ) ); \
|
||||
VD = _mm256_xor_si256( _mm256_set_epi64x( T0, T0, T0, T0 ), \
|
||||
_mm256_set_epi64x( CB5, CB5, CB5, CB5 ) ); \
|
||||
VE = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
|
||||
_mm256_set_epi64x( CB6, CB6, CB6, CB6 ) ); \
|
||||
VF = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
|
||||
_mm256_set_epi64x( CB7, CB7, CB7, CB7 ) ); \
|
||||
M[0x0] = mm256_bswap_64( *(buf+0) ); \
|
||||
M[0x1] = mm256_bswap_64( *(buf+1) ); \
|
||||
M[0x2] = mm256_bswap_64( *(buf+2) ); \
|
||||
M[0x3] = mm256_bswap_64( *(buf+3) ); \
|
||||
M[0x4] = mm256_bswap_64( *(buf+4) ); \
|
||||
M[0x5] = mm256_bswap_64( *(buf+5) ); \
|
||||
M[0x6] = mm256_bswap_64( *(buf+6) ); \
|
||||
M[0x7] = mm256_bswap_64( *(buf+7) ); \
|
||||
M[0x8] = mm256_bswap_64( *(buf+8) ); \
|
||||
M[0x9] = mm256_bswap_64( *(buf+9) ); \
|
||||
M[0xA] = mm256_bswap_64( *(buf+10) ); \
|
||||
M[0xB] = mm256_bswap_64( *(buf+11) ); \
|
||||
M[0xC] = mm256_bswap_64( *(buf+12) ); \
|
||||
M[0xD] = mm256_bswap_64( *(buf+13) ); \
|
||||
M[0xE] = mm256_bswap_64( *(buf+14) ); \
|
||||
M[0xF] = mm256_bswap_64( *(buf+15) ); \
|
||||
for (r = 0; r < 16; r ++) \
|
||||
ROUND_B_4WAY(r); \
|
||||
H0 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S0, V0 ), V8 ), H0 ); \
|
||||
H1 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S1, V1 ), V9 ), H1 ); \
|
||||
H2 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S2, V2 ), VA ), H2 ); \
|
||||
H3 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S3, V3 ), VB ), H3 ); \
|
||||
H4 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S0, V4 ), VC ), H4 ); \
|
||||
H5 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S1, V5 ), VD ), H5 ); \
|
||||
H6 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S2, V6 ), VE ), H6 ); \
|
||||
H7 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S3, V7 ), VF ), H7 ); \
|
||||
} while (0)
|
||||
|
||||
#else
|
||||
|
||||
//current impl
|
||||
|
||||
#define COMPRESS64_4WAY do { \
|
||||
__m256i M0, M1, M2, M3, M4, M5, M6, M7; \
|
||||
__m256i M8, M9, MA, MB, MC, MD, ME, MF; \
|
||||
__m256i V0, V1, V2, V3, V4, V5, V6, V7; \
|
||||
__m256i V8, V9, VA, VB, VC, VD, VE, VF; \
|
||||
V0 = H0; \
|
||||
V1 = H1; \
|
||||
V2 = H2; \
|
||||
V3 = H3; \
|
||||
V4 = H4; \
|
||||
V5 = H5; \
|
||||
V6 = H6; \
|
||||
V7 = H7; \
|
||||
V8 = _mm256_xor_si256( S0, _mm256_set_epi64x( CB0, CB0, CB0, CB0 ) ); \
|
||||
V9 = _mm256_xor_si256( S1, _mm256_set_epi64x( CB1, CB1, CB1, CB1 ) ); \
|
||||
VA = _mm256_xor_si256( S2, _mm256_set_epi64x( CB2, CB2, CB2, CB2 ) ); \
|
||||
VB = _mm256_xor_si256( S3, _mm256_set_epi64x( CB3, CB3, CB3, CB3 ) ); \
|
||||
VC = _mm256_xor_si256( _mm256_set_epi64x( T0, T0, T0, T0 ), \
|
||||
_mm256_set_epi64x( CB4, CB4, CB4, CB4 ) ); \
|
||||
VD = _mm256_xor_si256( _mm256_set_epi64x( T0, T0, T0, T0 ), \
|
||||
_mm256_set_epi64x( CB5, CB5, CB5, CB5 ) ); \
|
||||
VE = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
|
||||
_mm256_set_epi64x( CB6, CB6, CB6, CB6 ) ); \
|
||||
VF = _mm256_xor_si256( _mm256_set_epi64x( T1, T1, T1, T1 ), \
|
||||
_mm256_set_epi64x( CB7, CB7, CB7, CB7 ) ); \
|
||||
M0 = mm256_bswap_64( *(buf + 0) ); \
|
||||
M1 = mm256_bswap_64( *(buf + 1) ); \
|
||||
M2 = mm256_bswap_64( *(buf + 2) ); \
|
||||
M3 = mm256_bswap_64( *(buf + 3) ); \
|
||||
M4 = mm256_bswap_64( *(buf + 4) ); \
|
||||
M5 = mm256_bswap_64( *(buf + 5) ); \
|
||||
M6 = mm256_bswap_64( *(buf + 6) ); \
|
||||
M7 = mm256_bswap_64( *(buf + 7) ); \
|
||||
M8 = mm256_bswap_64( *(buf + 8) ); \
|
||||
M9 = mm256_bswap_64( *(buf + 9) ); \
|
||||
MA = mm256_bswap_64( *(buf + 10) ); \
|
||||
MB = mm256_bswap_64( *(buf + 11) ); \
|
||||
MC = mm256_bswap_64( *(buf + 12) ); \
|
||||
MD = mm256_bswap_64( *(buf + 13) ); \
|
||||
ME = mm256_bswap_64( *(buf + 14) ); \
|
||||
MF = mm256_bswap_64( *(buf + 15) ); \
|
||||
ROUND_B_4WAY(0); \
|
||||
ROUND_B_4WAY(1); \
|
||||
ROUND_B_4WAY(2); \
|
||||
ROUND_B_4WAY(3); \
|
||||
ROUND_B_4WAY(4); \
|
||||
ROUND_B_4WAY(5); \
|
||||
ROUND_B_4WAY(6); \
|
||||
ROUND_B_4WAY(7); \
|
||||
ROUND_B_4WAY(8); \
|
||||
ROUND_B_4WAY(9); \
|
||||
ROUND_B_4WAY(0); \
|
||||
ROUND_B_4WAY(1); \
|
||||
ROUND_B_4WAY(2); \
|
||||
ROUND_B_4WAY(3); \
|
||||
ROUND_B_4WAY(4); \
|
||||
ROUND_B_4WAY(5); \
|
||||
H0 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S0, V0 ), V8 ), H0 ); \
|
||||
H1 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S1, V1 ), V9 ), H1 ); \
|
||||
H2 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S2, V2 ), VA ), H2 ); \
|
||||
H3 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S3, V3 ), VB ), H3 ); \
|
||||
H4 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S0, V4 ), VC ), H4 ); \
|
||||
H5 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S1, V5 ), VD ), H5 ); \
|
||||
H6 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S2, V6 ), VE ), H6 ); \
|
||||
H7 = _mm256_xor_si256( _mm256_xor_si256( \
|
||||
_mm256_xor_si256( S3, V7 ), VF ), H7 ); \
|
||||
} while (0)
|
||||
|
||||
#endif
|
||||
|
||||
static const sph_u64 salt_zero_big[4] = { 0, 0, 0, 0 };
|
||||
|
||||
static void
|
||||
blake64_4way_init( blake_4way_big_context *sc, const sph_u64 *iv,
|
||||
const sph_u64 *salt )
|
||||
{
|
||||
int i;
|
||||
for ( i = 0; i < 8; i++ )
|
||||
sc->H[i] = _mm256_set1_epi64x( iv[i] );
|
||||
for ( i = 0; i < 4; i++ )
|
||||
sc->S[i] = _mm256_set1_epi64x( salt[i] );
|
||||
sc->T0 = sc->T1 = 0;
|
||||
sc->ptr = 0;
|
||||
}
|
||||
|
||||
static void
|
||||
blake64_4way( blake_4way_big_context *sc, const void *data, size_t len)
|
||||
{
|
||||
__m256i *vdata = (__m256i*)data;
|
||||
__m256i *buf;
|
||||
size_t ptr;
|
||||
DECL_STATE64_4WAY
|
||||
|
||||
const int buf_size = 128; // sizeof/8
|
||||
|
||||
buf = sc->buf;
|
||||
ptr = sc->ptr;
|
||||
if ( len < (buf_size - ptr) )
|
||||
{
|
||||
memcpy_256( buf + (ptr>>3), vdata, len>>3 );
|
||||
ptr += len;
|
||||
sc->ptr = ptr;
|
||||
return;
|
||||
}
|
||||
|
||||
READ_STATE64_4WAY(sc);
|
||||
while ( len > 0 )
|
||||
{
|
||||
size_t clen;
|
||||
|
||||
clen = buf_size - ptr;
|
||||
if ( clen > len )
|
||||
clen = len;
|
||||
memcpy_256( buf + (ptr>>3), vdata, clen>>3 );
|
||||
ptr += clen;
|
||||
vdata = vdata + (clen>>3);
|
||||
len -= clen;
|
||||
if (ptr == buf_size )
|
||||
{
|
||||
if ((T0 = SPH_T64(T0 + 1024)) < 1024)
|
||||
T1 = SPH_T64(T1 + 1);
|
||||
COMPRESS64_4WAY;
|
||||
ptr = 0;
|
||||
}
|
||||
}
|
||||
WRITE_STATE64_4WAY(sc);
|
||||
sc->ptr = ptr;
|
||||
}
|
||||
|
||||
static void
|
||||
blake64_4way_close( blake_4way_big_context *sc,
|
||||
unsigned ub, unsigned n, void *dst, size_t out_size_w64)
|
||||
{
|
||||
// union {
|
||||
__m256i buf[16];
|
||||
// sph_u64 dummy;
|
||||
// } u;
|
||||
size_t ptr, k;
|
||||
unsigned bit_len;
|
||||
uint64_t z, zz;
|
||||
sph_u64 th, tl;
|
||||
__m256i *out;
|
||||
|
||||
ptr = sc->ptr;
|
||||
bit_len = ((unsigned)ptr << 3);
|
||||
z = 0x80 >> n;
|
||||
zz = ((ub & -z) | z) & 0xFF;
|
||||
buf[ptr>>3] = _mm256_set_epi64x( zz, zz, zz, zz );
|
||||
tl = sc->T0 + bit_len;
|
||||
th = sc->T1;
|
||||
if (ptr == 0 )
|
||||
{
|
||||
sc->T0 = SPH_C64(0xFFFFFFFFFFFFFC00ULL);
|
||||
sc->T1 = SPH_C64(0xFFFFFFFFFFFFFFFFULL);
|
||||
}
|
||||
else if ( sc->T0 == 0 )
|
||||
{
|
||||
sc->T0 = SPH_C64(0xFFFFFFFFFFFFFC00ULL) + bit_len;
|
||||
sc->T1 = SPH_T64(sc->T1 - 1);
|
||||
}
|
||||
else
|
||||
{
|
||||
sc->T0 -= 1024 - bit_len;
|
||||
}
|
||||
if ( ptr <= 104 )
|
||||
{
|
||||
memset_zero_256( buf + (ptr>>3) + 1, (104-ptr) >> 3 );
|
||||
if ( out_size_w64 == 8 )
|
||||
buf[(104>>3)] = _mm256_or_si256( buf[(104>>3)],
|
||||
_mm256_set1_epi64x( 0x0100000000000000ULL ) );
|
||||
*(buf+(112>>3)) = mm256_bswap_64(
|
||||
_mm256_set_epi64x( th, th, th, th ) );
|
||||
*(buf+(120>>3)) = mm256_bswap_64(
|
||||
_mm256_set_epi64x( tl, tl, tl, tl ) );
|
||||
|
||||
blake64_4way( sc, buf + (ptr>>3), 128 - ptr );
|
||||
}
|
||||
else
|
||||
{
|
||||
memset_zero_256( buf + (ptr>>3) + 1, (120 - ptr) >> 3 );
|
||||
|
||||
blake64_4way( sc, buf + (ptr>>3), 128 - ptr );
|
||||
sc->T0 = SPH_C64(0xFFFFFFFFFFFFFC00ULL);
|
||||
sc->T1 = SPH_C64(0xFFFFFFFFFFFFFFFFULL);
|
||||
memset_zero_256( buf, 112>>3 );
|
||||
if ( out_size_w64 == 8 )
|
||||
buf[104>>3] = _mm256_set1_epi64x( 0x0100000000000000ULL );
|
||||
*(buf+(112>>3)) = mm256_bswap_64(
|
||||
_mm256_set_epi64x( th, th, th, th ) );
|
||||
*(buf+(120>>3)) = mm256_bswap_64(
|
||||
_mm256_set_epi64x( tl, tl, tl, tl ) );
|
||||
|
||||
blake64_4way( sc, buf, 128 );
|
||||
}
|
||||
out = (__m256i*)dst;
|
||||
for ( k = 0; k < out_size_w64; k++ )
|
||||
out[k] = mm256_bswap_64( sc->H[k] );
|
||||
}
|
||||
|
||||
void
|
||||
blake512_4way_init(void *cc)
|
||||
{
|
||||
blake64_4way_init(cc, IV512, salt_zero_big);
|
||||
}
|
||||
|
||||
void
|
||||
blake512_4way(void *cc, const void *data, size_t len)
|
||||
{
|
||||
blake64_4way(cc, data, len);
|
||||
}
|
||||
|
||||
void
|
||||
blake512_4way_close(void *cc, void *dst)
|
||||
{
|
||||
blake512_4way_addbits_and_close(cc, 0, 0, dst);
|
||||
}
|
||||
|
||||
void
|
||||
blake512_4way_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
|
||||
{
|
||||
blake64_4way_close(cc, ub, n, dst, 8);
|
||||
}
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif
|
||||
@@ -17,7 +17,7 @@ void blakecoin_4way_hash(void *state, const void *input)
|
||||
blake256r8_4way( &ctx, input + (64<<2), 16 );
|
||||
blake256r8_4way_close( &ctx, vhash );
|
||||
|
||||
mm_deinterleave_4x32( state, state+32, state+64, state+96, vhash, 256 );
|
||||
mm128_deinterleave_4x32( state, state+32, state+64, state+96, vhash, 256 );
|
||||
}
|
||||
|
||||
int scanhash_blakecoin_4way( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
@@ -37,7 +37,7 @@ int scanhash_blakecoin_4way( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
HTarget = 0x7f;
|
||||
|
||||
swab32_array( edata, pdata, 20 );
|
||||
mm_interleave_4x32( vdata, edata, edata, edata, edata, 640 );
|
||||
mm128_interleave_4x32( vdata, edata, edata, edata, edata, 640 );
|
||||
blake256r8_4way_init( &blakecoin_4w_ctx );
|
||||
blake256r8_4way( &blakecoin_4w_ctx, vdata, 64 );
|
||||
|
||||
|
||||
@@ -22,7 +22,7 @@ bool register_vanilla_algo( algo_gate_t* gate )
|
||||
gate->scanhash = (void*)&scanhash_blakecoin;
|
||||
gate->hash = (void*)&blakecoinhash;
|
||||
#endif
|
||||
gate->optimizations = AVX_OPT | AVX2_OPT;
|
||||
gate->optimizations = SSE42_OPT | AVX2_OPT;
|
||||
gate->get_max64 = (void*)&blakecoin_get_max64;
|
||||
return true;
|
||||
}
|
||||
|
||||
@@ -4,7 +4,7 @@
|
||||
#include "algo-gate-api.h"
|
||||
#include <stdint.h>
|
||||
|
||||
#if defined(__AVX__)
|
||||
#if defined(__SSE4_2__)
|
||||
#define BLAKECOIN_4WAY
|
||||
#endif
|
||||
#if defined(__AVX2__)
|
||||
|
||||
@@ -23,7 +23,7 @@ void decred_hash_4way( void *state, const void *input )
|
||||
memcpy( &ctx, &blake_mid, sizeof(blake_mid) );
|
||||
blake256_4way( &ctx, tail, tail_len );
|
||||
blake256_4way_close( &ctx, vhash );
|
||||
mm_deinterleave_4x32( state, state+32, state+64, state+96, vhash, 256 );
|
||||
mm128_deinterleave_4x32( state, state+32, state+64, state+96, vhash, 256 );
|
||||
}
|
||||
|
||||
int scanhash_decred_4way( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
@@ -44,7 +44,7 @@ int scanhash_decred_4way( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
memcpy( edata, pdata, 180 );
|
||||
|
||||
// use the old way until new way updated for size.
|
||||
mm_interleave_4x32x( vdata, edata, edata, edata, edata, 180*8 );
|
||||
mm128_interleave_4x32x( vdata, edata, edata, edata, edata, 180*8 );
|
||||
|
||||
blake256_4way_init( &blake_mid );
|
||||
blake256_4way( &blake_mid, vdata, DECRED_MIDSTATE_LEN );
|
||||
|
||||
@@ -140,6 +140,7 @@ bool decred_ready_to_mine( struct work* work, struct stratum_ctx* stratum,
|
||||
return true;
|
||||
}
|
||||
|
||||
int decred_get_work_data_size() { return DECRED_DATA_SIZE; }
|
||||
|
||||
bool register_decred_algo( algo_gate_t* gate )
|
||||
{
|
||||
@@ -154,7 +155,7 @@ bool register_decred_algo( algo_gate_t* gate )
|
||||
gate->optimizations = AVX2_OPT;
|
||||
gate->get_nonceptr = (void*)&decred_get_nonceptr;
|
||||
gate->get_max64 = (void*)&get_max64_0x3fffffLL;
|
||||
gate->display_extra_data = (void*)&decred_decode_extradata;
|
||||
gate->decode_extra_data = (void*)&decred_decode_extradata;
|
||||
gate->build_stratum_request = (void*)&decred_be_build_stratum_request;
|
||||
gate->work_decode = (void*)&std_be_work_decode;
|
||||
gate->submit_getwork_result = (void*)&std_be_submit_getwork_result;
|
||||
@@ -163,7 +164,7 @@ bool register_decred_algo( algo_gate_t* gate )
|
||||
gate->nbits_index = DECRED_NBITS_INDEX;
|
||||
gate->ntime_index = DECRED_NTIME_INDEX;
|
||||
gate->nonce_index = DECRED_NONCE_INDEX;
|
||||
gate->work_data_size = DECRED_DATA_SIZE;
|
||||
gate->get_work_data_size = (void*)&decred_get_work_data_size;
|
||||
gate->work_cmp_size = DECRED_WORK_COMPARE_SIZE;
|
||||
allow_mininginfo = false;
|
||||
have_gbt = false;
|
||||
|
||||
@@ -18,7 +18,7 @@
|
||||
// uint64_t *hashes_done );
|
||||
#endif
|
||||
|
||||
#if defined(__AVX2__)
|
||||
#if defined(__SSE4_2__)
|
||||
#define DECRED_4WAY
|
||||
#endif
|
||||
|
||||
|
||||
@@ -268,7 +268,7 @@ bool register_decred_algo( algo_gate_t* gate )
|
||||
gate->hash = (void*)&decred_hash;
|
||||
gate->get_nonceptr = (void*)&decred_get_nonceptr;
|
||||
gate->get_max64 = (void*)&get_max64_0x3fffffLL;
|
||||
gate->display_extra_data = (void*)&decred_decode_extradata;
|
||||
gate->decode_extra_data = (void*)&decred_decode_extradata;
|
||||
gate->build_stratum_request = (void*)&decred_be_build_stratum_request;
|
||||
gate->work_decode = (void*)&std_be_work_decode;
|
||||
gate->submit_getwork_result = (void*)&std_be_submit_getwork_result;
|
||||
|
||||
@@ -41,15 +41,18 @@ extern "C"{
|
||||
#endif
|
||||
|
||||
#include <stddef.h>
|
||||
#ifdef __AVX2__
|
||||
|
||||
#include "algo/sha/sph_types.h"
|
||||
#include "avxdefs.h"
|
||||
#include "simd-utils.h"
|
||||
|
||||
#define SPH_SIZE_bmw256 256
|
||||
|
||||
#define SPH_SIZE_bmw512 512
|
||||
|
||||
#if defined(__SSE2__)
|
||||
|
||||
// BMW-256 4 way 32
|
||||
|
||||
typedef struct {
|
||||
__m128i buf[64];
|
||||
__m128i H[16];
|
||||
@@ -59,6 +62,60 @@ typedef struct {
|
||||
|
||||
typedef bmw_4way_small_context bmw256_4way_context;
|
||||
|
||||
void bmw256_4way_init(void *cc);
|
||||
|
||||
void bmw256_4way(void *cc, const void *data, size_t len);
|
||||
|
||||
void bmw256_4way_close(void *cc, void *dst);
|
||||
|
||||
void bmw256_4way_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
#endif // __SSE2__
|
||||
|
||||
#if defined(__AVX2__)
|
||||
|
||||
// BMW-256 8 way 32
|
||||
|
||||
typedef struct {
|
||||
__m256i buf[64];
|
||||
__m256i H[16];
|
||||
size_t ptr;
|
||||
uint32_t bit_count; // assume bit_count fits in 32 bits
|
||||
} bmw_8way_small_context __attribute__ ((aligned (64)));
|
||||
|
||||
typedef bmw_8way_small_context bmw256_8way_context;
|
||||
|
||||
void bmw256_8way_init( bmw256_8way_context *ctx );
|
||||
void bmw256_8way( bmw256_8way_context *ctx, const void *data, size_t len );
|
||||
void bmw256_8way_close( bmw256_8way_context *ctx, void *dst );
|
||||
|
||||
#endif
|
||||
|
||||
|
||||
#if defined(__SSE2__)
|
||||
|
||||
// BMW-512 2 way 64
|
||||
|
||||
typedef struct {
|
||||
__m128i buf[16];
|
||||
__m128i H[16];
|
||||
size_t ptr;
|
||||
uint64_t bit_count;
|
||||
} bmw_2way_big_context __attribute__ ((aligned (64)));
|
||||
|
||||
typedef bmw_2way_big_context bmw512_2way_context;
|
||||
|
||||
void bmw512_2way_init( bmw512_2way_context *ctx );
|
||||
void bmw512_2way( bmw512_2way_context *ctx, const void *data, size_t len );
|
||||
void bmw512_2way_close( bmw512_2way_context *ctx, void *dst );
|
||||
|
||||
#endif // __SSE2__
|
||||
|
||||
#if defined(__AVX2__)
|
||||
|
||||
// BMW-512 4 way 64
|
||||
|
||||
typedef struct {
|
||||
__m256i buf[16];
|
||||
__m256i H[16];
|
||||
@@ -68,14 +125,6 @@ typedef struct {
|
||||
|
||||
typedef bmw_4way_big_context bmw512_4way_context;
|
||||
|
||||
void bmw256_4way_init(void *cc);
|
||||
|
||||
void bmw256_4way(void *cc, const void *data, size_t len);
|
||||
|
||||
void bmw256_4way_close(void *cc, void *dst);
|
||||
|
||||
void bmw256_4way_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
void bmw512_4way_init(void *cc);
|
||||
|
||||
@@ -86,10 +135,10 @@ void bmw512_4way_close(void *cc, void *dst);
|
||||
void bmw512_4way_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
#endif
|
||||
#endif // __AVX2__
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif
|
||||
#endif // BMW_HASH_H__
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
File diff suppressed because it is too large
Load Diff
@@ -325,7 +325,7 @@ int scanhash_cryptolight(int thr_id, struct work *work,
|
||||
|
||||
struct cryptonight_ctx *ctx = (struct cryptonight_ctx*)malloc(sizeof(struct cryptonight_ctx));
|
||||
|
||||
#ifndef NO_AES_NI
|
||||
#if defined(__AES__)
|
||||
do {
|
||||
*nonceptr = ++n;
|
||||
cryptolight_hash_ctx_aes_ni(hash, pdata, 76, ctx);
|
||||
|
||||
@@ -1,14 +1,11 @@
|
||||
#if defined(__AES__)
|
||||
|
||||
#include <x86intrin.h>
|
||||
#include <memory.h>
|
||||
#include "cryptonight.h"
|
||||
#include "miner.h"
|
||||
#include "crypto/c_keccak.h"
|
||||
#include <immintrin.h>
|
||||
//#include "avxdefs.h"
|
||||
|
||||
void aesni_parallel_noxor(uint8_t *long_state, uint8_t *text, uint8_t *ExpandedKey);
|
||||
void aesni_parallel_xor(uint8_t *text, uint8_t *ExpandedKey, uint8_t *long_state);
|
||||
void that_fucking_loop(uint8_t a[16], uint8_t b[16], uint8_t *long_state);
|
||||
|
||||
static inline void ExpandAESKey256_sub1(__m128i *tmp1, __m128i *tmp2)
|
||||
{
|
||||
@@ -25,7 +22,6 @@ static inline void ExpandAESKey256_sub1(__m128i *tmp1, __m128i *tmp2)
|
||||
|
||||
static inline void ExpandAESKey256_sub2(__m128i *tmp1, __m128i *tmp3)
|
||||
{
|
||||
#ifndef NO_AES_NI
|
||||
__m128i tmp2, tmp4;
|
||||
|
||||
tmp4 = _mm_aeskeygenassist_si128(*tmp1, 0x00);
|
||||
@@ -37,14 +33,12 @@ static inline void ExpandAESKey256_sub2(__m128i *tmp1, __m128i *tmp3)
|
||||
tmp4 = _mm_slli_si128(tmp4, 0x04);
|
||||
*tmp3 = _mm_xor_si128(*tmp3, tmp4);
|
||||
*tmp3 = _mm_xor_si128(*tmp3, tmp2);
|
||||
#endif
|
||||
}
|
||||
|
||||
// Special thanks to Intel for helping me
|
||||
// with ExpandAESKey256() and its subroutines
|
||||
static inline void ExpandAESKey256(char *keybuf)
|
||||
{
|
||||
#ifndef NO_AES_NI
|
||||
__m128i tmp1, tmp2, tmp3, *keys;
|
||||
|
||||
keys = (__m128i *)keybuf;
|
||||
@@ -91,7 +85,6 @@ static inline void ExpandAESKey256(char *keybuf)
|
||||
tmp2 = _mm_aeskeygenassist_si128(tmp3, 0x40);
|
||||
ExpandAESKey256_sub1(&tmp1, &tmp2);
|
||||
keys[14] = tmp1;
|
||||
#endif
|
||||
}
|
||||
|
||||
// align to 64 byte cache line
|
||||
@@ -109,13 +102,19 @@ static __thread cryptonight_ctx ctx;
|
||||
|
||||
void cryptonight_hash_aes( void *restrict output, const void *input, int len )
|
||||
{
|
||||
#ifndef NO_AES_NI
|
||||
|
||||
uint8_t ExpandedKey[256] __attribute__((aligned(64)));
|
||||
__m128i *longoutput, *expkey, *xmminput;
|
||||
size_t i, j;
|
||||
|
||||
keccak( (const uint8_t*)input, 76, (char*)&ctx.state.hs.b, 200 );
|
||||
|
||||
if ( cryptonightV7 && len < 43 )
|
||||
return;
|
||||
|
||||
const uint64_t tweak = cryptonightV7
|
||||
? *((const uint64_t*) (((const uint8_t*)input) + 35))
|
||||
^ ctx.state.hs.w[24] : 0;
|
||||
|
||||
memcpy( ExpandedKey, ctx.state.hs.b, AES_KEY_SIZE );
|
||||
ExpandAESKey256( ExpandedKey );
|
||||
memcpy( ctx.text, ctx.state.init, INIT_SIZE_BYTE );
|
||||
@@ -214,7 +213,15 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
|
||||
_mm_store_si128( (__m128i*)c, c_x );
|
||||
b_x = _mm_xor_si128( b_x, c_x );
|
||||
nextblock = (uint64_t *)&ctx.long_state[c[0] & 0x1FFFF0];
|
||||
_mm_store_si128( lsa, b_x );
|
||||
_mm_store_si128( lsa, b_x );
|
||||
|
||||
if ( cryptonightV7 )
|
||||
{
|
||||
const uint8_t tmp = ( (const uint8_t*)(lsa) )[11];
|
||||
const uint8_t index = ( ( (tmp >> 3) & 6 ) | (tmp & 1) ) << 1;
|
||||
((uint8_t*)(lsa))[11] = tmp ^ ( ( 0x75310 >> index) & 0x30 );
|
||||
}
|
||||
|
||||
b[0] = nextblock[0];
|
||||
b[1] = nextblock[1];
|
||||
|
||||
@@ -227,10 +234,14 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
|
||||
: "cc" );
|
||||
|
||||
b_x = c_x;
|
||||
nextblock[0] = a[0] + hi;
|
||||
nextblock[1] = a[1] + lo;
|
||||
a[0] = b[0] ^ nextblock[0];
|
||||
a[1] = b[1] ^ nextblock[1];
|
||||
|
||||
a[0] += hi;
|
||||
a[1] += lo;
|
||||
nextblock[0] = a[0];
|
||||
nextblock[1] = cryptonightV7 ? a[1] ^ tweak : a[1];
|
||||
a[0] ^= b[0];
|
||||
a[1] ^= b[1];
|
||||
|
||||
lsa = (__m128i*)&ctx.long_state[ a[0] & 0x1FFFF0 ];
|
||||
a_x = _mm_load_si128( (__m128i*)a );
|
||||
c_x = _mm_load_si128( lsa );
|
||||
@@ -241,6 +252,14 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
|
||||
b_x = _mm_xor_si128( b_x, c_x );
|
||||
nextblock = (uint64_t *)&ctx.long_state[c[0] & 0x1FFFF0];
|
||||
_mm_store_si128( lsa, b_x );
|
||||
|
||||
if ( cryptonightV7 )
|
||||
{
|
||||
const uint8_t tmp = ( (const uint8_t*)(lsa) )[11];
|
||||
const uint8_t index = ( ( (tmp >> 3) & 6 ) | (tmp & 1) ) << 1;
|
||||
((uint8_t*)(lsa))[11] = tmp ^ ( ( 0x75310 >> index) & 0x30 );
|
||||
}
|
||||
|
||||
b[0] = nextblock[0];
|
||||
b[1] = nextblock[1];
|
||||
|
||||
@@ -251,8 +270,12 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
|
||||
"rm" ( b[0] )
|
||||
: "cc" );
|
||||
|
||||
nextblock[0] = a[0] + hi;
|
||||
nextblock[1] = a[1] + lo;
|
||||
a[0] += hi;
|
||||
a[1] += lo;
|
||||
nextblock[0] = a[0];
|
||||
nextblock[1] = cryptonightV7 ? a[1] ^ tweak : a[1];
|
||||
a[0] ^= b[0];
|
||||
a[1] ^= b[1];
|
||||
|
||||
memcpy( ExpandedKey, &ctx.state.hs.b[32], AES_KEY_SIZE );
|
||||
ExpandAESKey256( ExpandedKey );
|
||||
@@ -330,5 +353,5 @@ void cryptonight_hash_aes( void *restrict output, const void *input, int len )
|
||||
keccakf( (uint64_t*)&ctx.state.hs.w, 24 );
|
||||
extra_hashes[ctx.state.hs.b[0] & 3](&ctx.state, 200, output);
|
||||
|
||||
#endif
|
||||
}
|
||||
#endif
|
||||
|
||||
@@ -7,11 +7,11 @@
|
||||
#include "cpuminer-config.h"
|
||||
#include "algo-gate-api.h"
|
||||
|
||||
#ifndef NO_AES_NI
|
||||
#if defined(__AES__)
|
||||
#include "algo/groestl/aes_ni/hash-groestl256.h"
|
||||
#endif
|
||||
|
||||
#else
|
||||
#include "crypto/c_groestl.h"
|
||||
#endif
|
||||
#include "crypto/c_blake256.h"
|
||||
#include "crypto/c_jh.h"
|
||||
#include "crypto/c_skein.h"
|
||||
@@ -30,12 +30,12 @@ void do_blake_hash(const void* input, size_t len, char* output) {
|
||||
}
|
||||
|
||||
void do_groestl_hash(const void* input, size_t len, char* output) {
|
||||
#ifdef NO_AES_NI
|
||||
groestl(input, len * 8, (uint8_t*)output);
|
||||
#else
|
||||
#if defined(__AES__)
|
||||
hashState_groestl256 ctx;
|
||||
init_groestl256( &ctx, 32 );
|
||||
update_and_final_groestl256( &ctx, output, input, len * 8 );
|
||||
#else
|
||||
groestl(input, len * 8, (uint8_t*)output);
|
||||
#endif
|
||||
}
|
||||
|
||||
@@ -52,23 +52,24 @@ void (* const extra_hashes[4])( const void *, size_t, char *) =
|
||||
|
||||
void cryptonight_hash( void *restrict output, const void *input, int len )
|
||||
{
|
||||
|
||||
#ifdef NO_AES_NI
|
||||
cryptonight_hash_ctx ( output, input, len );
|
||||
#else
|
||||
#if defined(__AES__)
|
||||
cryptonight_hash_aes( output, input, len );
|
||||
#else
|
||||
cryptonight_hash_ctx ( output, input, len );
|
||||
#endif
|
||||
}
|
||||
|
||||
void cryptonight_hash_suw( void *restrict output, const void *input )
|
||||
{
|
||||
#ifdef NO_AES_NI
|
||||
cryptonight_hash_ctx ( output, input, 76 );
|
||||
#else
|
||||
#if defined(__AES__)
|
||||
cryptonight_hash_aes( output, input, 76 );
|
||||
#else
|
||||
cryptonight_hash_ctx ( output, input, 76 );
|
||||
#endif
|
||||
}
|
||||
|
||||
bool cryptonightV7 = false;
|
||||
|
||||
int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
uint64_t *hashes_done )
|
||||
{
|
||||
@@ -80,6 +81,11 @@ int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
const uint32_t first_nonce = n + 1;
|
||||
const uint32_t Htarg = ptarget[7];
|
||||
uint32_t hash[32 / 4] __attribute__((aligned(32)));
|
||||
|
||||
// if ( ( cryptonightV7 && ( *(uint8_t*)pdata < 7 ) )
|
||||
// || ( !cryptonightV7 && ( *(uint8_t*)pdata == 7 ) ) )
|
||||
// applog(LOG_WARNING,"Cryptonight variant mismatch, shares may be rejected.");
|
||||
|
||||
do
|
||||
{
|
||||
*nonceptr = ++n;
|
||||
@@ -87,6 +93,7 @@ int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
if (unlikely( hash[7] < Htarg ))
|
||||
{
|
||||
*hashes_done = n - first_nonce + 1;
|
||||
// work_set_target_ratio( work, hash );
|
||||
return true;
|
||||
}
|
||||
} while (likely((n <= max_nonce && !work_restart[thr_id].restart)));
|
||||
@@ -97,6 +104,7 @@ int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
|
||||
bool register_cryptonight_algo( algo_gate_t* gate )
|
||||
{
|
||||
cryptonightV7 = false;
|
||||
register_json_rpc2( gate );
|
||||
gate->optimizations = SSE2_OPT | AES_OPT;
|
||||
gate->scanhash = (void*)&scanhash_cryptonight;
|
||||
@@ -106,3 +114,15 @@ bool register_cryptonight_algo( algo_gate_t* gate )
|
||||
return true;
|
||||
};
|
||||
|
||||
bool register_cryptonightv7_algo( algo_gate_t* gate )
|
||||
{
|
||||
cryptonightV7 = true;
|
||||
register_json_rpc2( gate );
|
||||
gate->optimizations = SSE2_OPT | AES_OPT;
|
||||
gate->scanhash = (void*)&scanhash_cryptonight;
|
||||
gate->hash = (void*)&cryptonight_hash;
|
||||
gate->hash_suw = (void*)&cryptonight_hash_suw;
|
||||
gate->get_max64 = (void*)&get_max64_0x40LL;
|
||||
return true;
|
||||
};
|
||||
|
||||
|
||||
@@ -20,8 +20,8 @@
|
||||
#include "crypto/c_jh.h"
|
||||
#include "crypto/c_skein.h"
|
||||
#include "crypto/int-util.h"
|
||||
#include "crypto/hash-ops.h"
|
||||
//#include "cryptonight.h"
|
||||
//#include "crypto/hash-ops.h"
|
||||
#include "cryptonight.h"
|
||||
|
||||
#if USE_INT128
|
||||
|
||||
@@ -51,6 +51,7 @@ typedef __uint128_t uint128_t;
|
||||
#define INIT_SIZE_BLK 8
|
||||
#define INIT_SIZE_BYTE (INIT_SIZE_BLK * AES_BLOCK_SIZE)
|
||||
|
||||
/*
|
||||
#pragma pack(push, 1)
|
||||
union cn_slow_hash_state {
|
||||
union hash_state hs;
|
||||
@@ -78,6 +79,7 @@ static void do_skein_hash(const void* input, size_t len, char* output) {
|
||||
int r = skein_hash(8 * HASH_SIZE, input, 8 * len, (uint8_t*)output);
|
||||
assert(likely(SKEIN_SUCCESS == r));
|
||||
}
|
||||
*/
|
||||
|
||||
extern int aesb_single_round(const uint8_t *in, uint8_t*out, const uint8_t *expandedKey);
|
||||
extern int aesb_pseudo_round_mut(uint8_t *val, uint8_t *expandedKey);
|
||||
@@ -120,9 +122,11 @@ static uint64_t mul128(uint64_t multiplier, uint64_t multiplicand, uint64_t* pro
|
||||
extern uint64_t mul128(uint64_t multiplier, uint64_t multiplicand, uint64_t* product_hi);
|
||||
#endif
|
||||
|
||||
/*
|
||||
static void (* const extra_hashes[4])(const void *, size_t, char *) = {
|
||||
do_blake_hash, do_groestl_hash, do_jh_hash, do_skein_hash
|
||||
};
|
||||
*/
|
||||
|
||||
static inline size_t e2i(const uint8_t* a) {
|
||||
#if !LITE
|
||||
@@ -132,14 +136,16 @@ static inline size_t e2i(const uint8_t* a) {
|
||||
#endif
|
||||
}
|
||||
|
||||
static inline void mul_sum_xor_dst(const uint8_t* a, uint8_t* c, uint8_t* dst) {
|
||||
static inline void mul_sum_xor_dst( const uint8_t* a, uint8_t* c, uint8_t* dst,
|
||||
const uint64_t tweak )
|
||||
{
|
||||
uint64_t hi, lo = mul128(((uint64_t*) a)[0], ((uint64_t*) dst)[0], &hi) + ((uint64_t*) c)[1];
|
||||
hi += ((uint64_t*) c)[0];
|
||||
|
||||
((uint64_t*) c)[0] = ((uint64_t*) dst)[0] ^ hi;
|
||||
((uint64_t*) c)[1] = ((uint64_t*) dst)[1] ^ lo;
|
||||
((uint64_t*) dst)[0] = hi;
|
||||
((uint64_t*) dst)[1] = lo;
|
||||
((uint64_t*) dst)[1] = cryptonightV7 ? lo ^ tweak : lo;
|
||||
}
|
||||
|
||||
static inline void xor_blocks(uint8_t* a, const uint8_t* b) {
|
||||
@@ -174,8 +180,16 @@ static __thread cryptonight_ctx ctx;
|
||||
|
||||
void cryptonight_hash_ctx(void* output, const void* input, int len)
|
||||
{
|
||||
hash_process(&ctx.state.hs, (const uint8_t*) input, len);
|
||||
ctx.aes_ctx = (oaes_ctx*) oaes_alloc();
|
||||
// hash_process(&ctx.state.hs, (const uint8_t*) input, len);
|
||||
keccak( (const uint8_t*)input, 76, (char*)&ctx.state.hs.b, 200 );
|
||||
|
||||
if ( cryptonightV7 && len < 43 )
|
||||
return;
|
||||
const uint64_t tweak = cryptonightV7
|
||||
? *((const uint64_t*) (((const uint8_t*)input) + 35))
|
||||
^ ctx.state.hs.w[24] : 0;
|
||||
|
||||
ctx.aes_ctx = (oaes_ctx*) oaes_alloc();
|
||||
|
||||
__builtin_prefetch( ctx.text, 0, 3 );
|
||||
__builtin_prefetch( ctx.text + 64, 0, 3 );
|
||||
@@ -211,23 +225,44 @@ void cryptonight_hash_ctx(void* output, const void* input, int len)
|
||||
xor_blocks_dst(&ctx.state.k[0], &ctx.state.k[32], ctx.a);
|
||||
xor_blocks_dst(&ctx.state.k[16], &ctx.state.k[48], ctx.b);
|
||||
|
||||
for (i = 0; likely(i < ITER / 4); ++i) {
|
||||
/* Dependency chain: address -> read value ------+
|
||||
* written value <-+ hard function (AES or MUL) <+
|
||||
* next address <-+
|
||||
*/
|
||||
/* Iteration 1 */
|
||||
j = e2i(ctx.a);
|
||||
aesb_single_round(&ctx.long_state[j], ctx.c, ctx.a);
|
||||
xor_blocks_dst(ctx.c, ctx.b, &ctx.long_state[j]);
|
||||
/* Iteration 2 */
|
||||
mul_sum_xor_dst(ctx.c, ctx.a, &ctx.long_state[e2i(ctx.c)]);
|
||||
/* Iteration 3 */
|
||||
j = e2i(ctx.a);
|
||||
aesb_single_round(&ctx.long_state[j], ctx.b, ctx.a);
|
||||
xor_blocks_dst(ctx.b, ctx.c, &ctx.long_state[j]);
|
||||
/* Iteration 4 */
|
||||
mul_sum_xor_dst(ctx.b, ctx.a, &ctx.long_state[e2i(ctx.b)]);
|
||||
for (i = 0; likely(i < ITER / 4); ++i)
|
||||
{
|
||||
/* Dependency chain: address -> read value ------+
|
||||
* written value <-+ hard function (AES or MUL) <+
|
||||
* next address <-+
|
||||
*/
|
||||
/* Iteration 1 */
|
||||
j = e2i(ctx.a);
|
||||
aesb_single_round(&ctx.long_state[j], ctx.c, ctx.a);
|
||||
xor_blocks_dst(ctx.c, ctx.b, &ctx.long_state[j]);
|
||||
|
||||
if ( cryptonightV7 )
|
||||
{
|
||||
uint8_t *lsa = (uint8_t*)&ctx.long_state[((uint64_t *)(ctx.a))[0] & 0x1FFFF0];
|
||||
const uint8_t tmp = lsa[11];
|
||||
const uint8_t index = ( ( (tmp >> 3) & 6 ) | (tmp & 1) ) << 1;
|
||||
lsa[11] = tmp ^ ( ( 0x75310 >> index) & 0x30 );
|
||||
}
|
||||
|
||||
/* Iteration 2 */
|
||||
mul_sum_xor_dst(ctx.c, ctx.a, &ctx.long_state[e2i(ctx.c)], tweak );
|
||||
|
||||
/* Iteration 3 */
|
||||
j = e2i(ctx.a);
|
||||
aesb_single_round(&ctx.long_state[j], ctx.b, ctx.a);
|
||||
xor_blocks_dst(ctx.b, ctx.c, &ctx.long_state[j]);
|
||||
|
||||
if ( cryptonightV7 )
|
||||
{
|
||||
uint8_t *lsa = (uint8_t*)&ctx.long_state[((uint64_t *)(ctx.a))[0] & 0x1FFFF0];
|
||||
const uint8_t tmp = lsa[11];
|
||||
const uint8_t index = ( ( (tmp >> 3) & 6 ) | (tmp & 1) ) << 1;
|
||||
lsa[11] = tmp ^ ( ( 0x75310 >> index) & 0x30 );
|
||||
}
|
||||
|
||||
/* Iteration 4 */
|
||||
mul_sum_xor_dst(ctx.b, ctx.a, &ctx.long_state[e2i(ctx.b)], tweak );
|
||||
|
||||
}
|
||||
|
||||
__builtin_prefetch( ctx.text, 0, 3 );
|
||||
@@ -266,7 +301,8 @@ void cryptonight_hash_ctx(void* output, const void* input, int len)
|
||||
aesb_pseudo_round_mut(&ctx.text[7 * AES_BLOCK_SIZE], ctx.aes_ctx->key->exp_data);
|
||||
}
|
||||
memcpy(ctx.state.init, ctx.text, INIT_SIZE_BYTE);
|
||||
hash_permutation(&ctx.state.hs);
|
||||
// hash_permutation(&ctx.state.hs);
|
||||
keccakf( (uint64_t*)&ctx.state.hs.w, 24 );
|
||||
/*memcpy(hash, &state, 32);*/
|
||||
extra_hashes[ctx.state.hs.b[0] & 3](&ctx.state, 200, output);
|
||||
oaes_free((OAES_CTX **) &ctx.aes_ctx);
|
||||
|
||||
@@ -45,5 +45,7 @@ int scanhash_cryptonight( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
|
||||
void cryptonight_hash_aes( void *restrict output, const void *input, int len );
|
||||
|
||||
extern bool cryptonightV7;
|
||||
|
||||
#endif
|
||||
|
||||
|
||||
@@ -7,6 +7,24 @@
|
||||
|
||||
// 2x128
|
||||
|
||||
// The result of hashing 10 rounds of initial data which consists of params
|
||||
// zero padded.
|
||||
static const uint64_t IV256[] =
|
||||
{
|
||||
0xCCD6F29FEA2BD4B4, 0x35481EAE63117E71, 0xE5D94E6322512D5B, 0xF4CC12BE7E624131,
|
||||
0x42AF2070C2D0B696, 0x3361DA8CD0720C35, 0x8EF8AD8328CCECA4, 0x40E5FBAB4680AC00,
|
||||
0x6107FBD5D89041C3, 0xF0B266796C859D41, 0x5FA2560309392549, 0x93CB628565C892FD,
|
||||
0x9E4B4E602AF2B5AE, 0x85254725774ABFDD, 0x4AB6AAD615815AEB, 0xD6032C0A9CDAF8AF
|
||||
};
|
||||
|
||||
static const uint64_t IV512[] =
|
||||
{
|
||||
0x50F494D42AEA2A61, 0x4167D83E2D538B8B, 0xC701CF8C3FEE2313, 0x50AC5695CC39968E,
|
||||
0xA647A8B34D42C787, 0x825B453797CF0BEF, 0xF22090C4EEF864D2, 0xA23911AED0E5CD33,
|
||||
0x148FE485FCD398D9, 0xB64445321B017BEF, 0x2FF5781C6A536159, 0x0DBADEA991FA7934,
|
||||
0xA5A70E75D65C8A2B, 0xBC796576B1C62456, 0xE7989AF11921C8F7, 0xD43E3B447795D246
|
||||
};
|
||||
|
||||
static void transform_2way( cube_2way_context *sp )
|
||||
{
|
||||
int r;
|
||||
@@ -45,10 +63,10 @@ static void transform_2way( cube_2way_context *sp )
|
||||
x1 = _mm256_xor_si256( x1, x5 );
|
||||
x2 = _mm256_xor_si256( x2, x6 );
|
||||
x3 = _mm256_xor_si256( x3, x7 );
|
||||
x4 = mm256_swap128_64( x4 );
|
||||
x5 = mm256_swap128_64( x5 );
|
||||
x6 = mm256_swap128_64( x6 );
|
||||
x7 = mm256_swap128_64( x7 );
|
||||
x4 = mm256_swap64_128( x4 );
|
||||
x5 = mm256_swap64_128( x5 );
|
||||
x6 = mm256_swap64_128( x6 );
|
||||
x7 = mm256_swap64_128( x7 );
|
||||
x4 = _mm256_add_epi32( x0, x4 );
|
||||
x5 = _mm256_add_epi32( x1, x5 );
|
||||
x6 = _mm256_add_epi32( x2, x6 );
|
||||
@@ -69,10 +87,10 @@ static void transform_2way( cube_2way_context *sp )
|
||||
x1 = _mm256_xor_si256( x1, x5 );
|
||||
x2 = _mm256_xor_si256( x2, x6 );
|
||||
x3 = _mm256_xor_si256( x3, x7 );
|
||||
x4 = mm256_swap64_32( x4 );
|
||||
x5 = mm256_swap64_32( x5 );
|
||||
x6 = mm256_swap64_32( x6 );
|
||||
x7 = mm256_swap64_32( x7 );
|
||||
x4 = mm256_swap32_64( x4 );
|
||||
x5 = mm256_swap32_64( x5 );
|
||||
x6 = mm256_swap32_64( x6 );
|
||||
x7 = mm256_swap32_64( x7 );
|
||||
}
|
||||
|
||||
_mm256_store_si256( (__m256i*)sp->h, x0 );
|
||||
@@ -86,44 +104,33 @@ static void transform_2way( cube_2way_context *sp )
|
||||
|
||||
}
|
||||
|
||||
cube_2way_context cube_2way_ctx_cache __attribute__ ((aligned (64)));
|
||||
|
||||
int cube_2way_reinit( cube_2way_context *sp )
|
||||
{
|
||||
memcpy( sp, &cube_2way_ctx_cache, sizeof(cube_2way_context) );
|
||||
return 0;
|
||||
|
||||
}
|
||||
|
||||
int cube_2way_init( cube_2way_context *sp, int hashbitlen, int rounds,
|
||||
int blockbytes )
|
||||
int blockbytes )
|
||||
{
|
||||
int i;
|
||||
const uint64_t* iv = hashbitlen == 512 ? IV512 : IV256;
|
||||
sp->hashlen = hashbitlen/128;
|
||||
sp->blocksize = blockbytes/16;
|
||||
sp->rounds = rounds;
|
||||
sp->pos = 0;
|
||||
|
||||
// all sizes of __m128i
|
||||
cube_2way_ctx_cache.hashlen = hashbitlen/128;
|
||||
cube_2way_ctx_cache.blocksize = blockbytes/16;
|
||||
cube_2way_ctx_cache.rounds = rounds;
|
||||
cube_2way_ctx_cache.pos = 0;
|
||||
__m256i* h = (__m256i*)sp->h;
|
||||
|
||||
for ( i = 0; i < 8; ++i )
|
||||
cube_2way_ctx_cache.h[i] = m256_zero;
|
||||
h[0] = _mm256_set_epi64x( iv[ 1], iv[ 0], iv[ 1], iv[ 0] );
|
||||
h[1] = _mm256_set_epi64x( iv[ 3], iv[ 2], iv[ 3], iv[ 2] );
|
||||
h[2] = _mm256_set_epi64x( iv[ 5], iv[ 4], iv[ 5], iv[ 4] );
|
||||
h[3] = _mm256_set_epi64x( iv[ 7], iv[ 6], iv[ 7], iv[ 6] );
|
||||
h[4] = _mm256_set_epi64x( iv[ 9], iv[ 8], iv[ 9], iv[ 8] );
|
||||
h[5] = _mm256_set_epi64x( iv[11], iv[10], iv[11], iv[10] );
|
||||
h[6] = _mm256_set_epi64x( iv[13], iv[12], iv[13], iv[12] );
|
||||
h[7] = _mm256_set_epi64x( iv[15], iv[14], iv[15], iv[14] );
|
||||
|
||||
cube_2way_ctx_cache.h[0] = _mm256_set_epi32(
|
||||
0, rounds, blockbytes, hashbitlen / 8,
|
||||
0, rounds, blockbytes, hashbitlen / 8 );
|
||||
|
||||
for ( i = 0; i < 10; ++i )
|
||||
transform_2way( &cube_2way_ctx_cache );
|
||||
|
||||
memcpy( sp, &cube_2way_ctx_cache, sizeof(cube_2way_context) );
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
int cube_2way_update( cube_2way_context *sp, const void *data, size_t size )
|
||||
{
|
||||
const int len = size / 16;
|
||||
const int len = size >> 4;
|
||||
const __m256i *in = (__m256i*)data;
|
||||
int i;
|
||||
|
||||
@@ -140,7 +147,6 @@ int cube_2way_update( cube_2way_context *sp, const void *data, size_t size )
|
||||
sp->pos = 0;
|
||||
}
|
||||
}
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
@@ -151,25 +157,22 @@ int cube_2way_close( cube_2way_context *sp, void *output )
|
||||
|
||||
// pos is zero for 64 byte data, 1 for 80 byte data.
|
||||
sp->h[ sp->pos ] = _mm256_xor_si256( sp->h[ sp->pos ],
|
||||
_mm256_set_epi8( 0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,0x80,
|
||||
0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,0x80 ) );
|
||||
_mm256_set_epi32( 0,0,0,0x80, 0,0,0,0x80 ) );
|
||||
transform_2way( sp );
|
||||
|
||||
sp->h[7] = _mm256_xor_si256( sp->h[7], _mm256_set_epi32( 1,0,0,0,
|
||||
1,0,0,0 ) );
|
||||
for ( i = 0; i < 10; ++i )
|
||||
transform_2way( &cube_2way_ctx_cache );
|
||||
sp->h[7] = _mm256_xor_si256( sp->h[7],
|
||||
_mm256_set_epi32( 1,0,0,0, 1,0,0,0 ) );
|
||||
|
||||
for ( i = 0; i < sp->hashlen; i++ )
|
||||
hash[i] = sp->h[i];
|
||||
for ( i = 0; i < 10; ++i ) transform_2way( sp );
|
||||
|
||||
for ( i = 0; i < sp->hashlen; i++ ) hash[i] = sp->h[i];
|
||||
return 0;
|
||||
}
|
||||
|
||||
int cube_2way_update_close( cube_2way_context *sp, void *output,
|
||||
const void *data, size_t size )
|
||||
{
|
||||
const int len = size / 16;
|
||||
const int len = size >> 4;
|
||||
const __m256i *in = (__m256i*)data;
|
||||
__m256i *hash = (__m256i*)output;
|
||||
int i;
|
||||
@@ -187,18 +190,15 @@ int cube_2way_update_close( cube_2way_context *sp, void *output,
|
||||
|
||||
// pos is zero for 64 byte data, 1 for 80 byte data.
|
||||
sp->h[ sp->pos ] = _mm256_xor_si256( sp->h[ sp->pos ],
|
||||
_mm256_set_epi8( 0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,0x80,
|
||||
0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,0x80 ) );
|
||||
_mm256_set_epi32( 0,0,0,0x80, 0,0,0,0x80 ) );
|
||||
transform_2way( sp );
|
||||
|
||||
sp->h[7] = _mm256_xor_si256( sp->h[7], _mm256_set_epi32( 1,0,0,0,
|
||||
1,0,0,0 ) );
|
||||
for ( i = 0; i < 10; ++i )
|
||||
transform_2way( &cube_2way_ctx_cache );
|
||||
|
||||
for ( i = 0; i < sp->hashlen; i++ )
|
||||
hash[i] = sp->h[i];
|
||||
for ( i = 0; i < 10; ++i ) transform_2way( sp );
|
||||
|
||||
for ( i = 0; i < sp->hashlen; i++ ) hash[i] = sp->h[i];
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
@@ -4,18 +4,18 @@
|
||||
#if defined(__AVX2__)
|
||||
|
||||
#include <stdint.h>
|
||||
#include "avxdefs.h"
|
||||
#include "simd-utils.h"
|
||||
|
||||
// 2x128, 2 way parallel SSE2
|
||||
|
||||
struct _cube_2way_context
|
||||
{
|
||||
__m256i h[8];
|
||||
int hashlen; // __m128i
|
||||
int rounds;
|
||||
int blocksize; // __m128i
|
||||
int pos; // number of __m128i read into x from current block
|
||||
__m256i h[8] __attribute__ ((aligned (64)));
|
||||
};
|
||||
} __attribute__ ((aligned (64)));
|
||||
|
||||
typedef struct _cube_2way_context cube_2way_context;
|
||||
|
||||
|
||||
@@ -13,7 +13,26 @@
|
||||
#include <stdbool.h>
|
||||
#include <unistd.h>
|
||||
#include <memory.h>
|
||||
#include "avxdefs.h"
|
||||
#include "simd-utils.h"
|
||||
#include <stdio.h>
|
||||
|
||||
// The result of hashing 10 rounds of initial data which is params and
|
||||
// mostly zeros.
|
||||
static const uint64_t IV256[] =
|
||||
{
|
||||
0xCCD6F29FEA2BD4B4, 0x35481EAE63117E71, 0xE5D94E6322512D5B, 0xF4CC12BE7E624131,
|
||||
0x42AF2070C2D0B696, 0x3361DA8CD0720C35, 0x8EF8AD8328CCECA4, 0x40E5FBAB4680AC00,
|
||||
0x6107FBD5D89041C3, 0xF0B266796C859D41, 0x5FA2560309392549, 0x93CB628565C892FD,
|
||||
0x9E4B4E602AF2B5AE, 0x85254725774ABFDD, 0x4AB6AAD615815AEB, 0xD6032C0A9CDAF8AF
|
||||
};
|
||||
|
||||
static const uint64_t IV512[] =
|
||||
{
|
||||
0x50F494D42AEA2A61, 0x4167D83E2D538B8B, 0xC701CF8C3FEE2313, 0x50AC5695CC39968E,
|
||||
0xA647A8B34D42C787, 0x825B453797CF0BEF, 0xF22090C4EEF864D2, 0xA23911AED0E5CD33,
|
||||
0x148FE485FCD398D9, 0xB64445321B017BEF, 0x2FF5781C6A536159, 0x0DBADEA991FA7934,
|
||||
0xA5A70E75D65C8A2B, 0xBC796576B1C62456, 0xE7989AF11921C8F7, 0xD43E3B447795D246
|
||||
};
|
||||
|
||||
static void transform( cubehashParam *sp )
|
||||
{
|
||||
@@ -22,7 +41,7 @@ static void transform( cubehashParam *sp )
|
||||
|
||||
#ifdef __AVX2__
|
||||
|
||||
__m256i x0, x1, x2, x3, y0, y1;
|
||||
register __m256i x0, x1, x2, x3, y0, y1;
|
||||
|
||||
x0 = _mm256_load_si256( (__m256i*)sp->x );
|
||||
x1 = _mm256_load_si256( (__m256i*)sp->x + 1 );
|
||||
@@ -33,20 +52,19 @@ static void transform( cubehashParam *sp )
|
||||
{
|
||||
x2 = _mm256_add_epi32( x0, x2 );
|
||||
x3 = _mm256_add_epi32( x1, x3 );
|
||||
y0 = x1;
|
||||
y1 = x0;
|
||||
x0 = _mm256_xor_si256( _mm256_slli_epi32( y0, 7 ),
|
||||
y0 = x0;
|
||||
x0 = _mm256_xor_si256( _mm256_slli_epi32( x1, 7 ),
|
||||
_mm256_srli_epi32( x1, 25 ) );
|
||||
x1 = _mm256_xor_si256( _mm256_slli_epi32( y0, 7 ),
|
||||
_mm256_srli_epi32( y0, 25 ) );
|
||||
x1 = _mm256_xor_si256( _mm256_slli_epi32( y1, 7 ),
|
||||
_mm256_srli_epi32( y1, 25 ) );
|
||||
x0 = _mm256_xor_si256( x0, x2 );
|
||||
x1 = _mm256_xor_si256( x1, x3 );
|
||||
x2 = _mm256_shuffle_epi32( x2, 0x4e );
|
||||
x3 = _mm256_shuffle_epi32( x3, 0x4e );
|
||||
x2 = _mm256_add_epi32( x0, x2 );
|
||||
x3 = _mm256_add_epi32( x1, x3 );
|
||||
y0 = _mm256_permute2f128_si256( x0, x0, 1 );
|
||||
y1 = _mm256_permute2f128_si256( x1, x1, 1 );
|
||||
y0 = _mm256_permute4x64_epi64( x0, 0x4e );
|
||||
y1 = _mm256_permute4x64_epi64( x1, 0x4e );
|
||||
x0 = _mm256_xor_si256( _mm256_slli_epi32( y0, 11 ),
|
||||
_mm256_srli_epi32( y0, 21 ) );
|
||||
x1 = _mm256_xor_si256( _mm256_slli_epi32( y1, 11 ),
|
||||
@@ -129,48 +147,37 @@ static void transform( cubehashParam *sp )
|
||||
#endif
|
||||
} // transform
|
||||
|
||||
// Cubehash context initializing is very expensive.
|
||||
// Cache the intial value for faster reinitializing.
|
||||
cubehashParam cube_ctx_cache __attribute__ ((aligned (64)));
|
||||
|
||||
int cubehashReinit( cubehashParam *sp )
|
||||
{
|
||||
memcpy( sp, &cube_ctx_cache, sizeof(cubehashParam) );
|
||||
return SUCCESS;
|
||||
|
||||
}
|
||||
|
||||
// Initialize the cache then copy to sp.
|
||||
int cubehashInit(cubehashParam *sp, int hashbitlen, int rounds, int blockbytes)
|
||||
{
|
||||
int i;
|
||||
const uint64_t* iv = hashbitlen == 512 ? IV512 : IV256;
|
||||
sp->hashlen = hashbitlen/128;
|
||||
sp->blocksize = blockbytes/16;
|
||||
sp->rounds = rounds;
|
||||
sp->pos = 0;
|
||||
|
||||
#if defined(__AVX2__)
|
||||
|
||||
if ( hashbitlen < 8 ) return BAD_HASHBITLEN;
|
||||
if ( hashbitlen > 512 ) return BAD_HASHBITLEN;
|
||||
if ( hashbitlen != 8 * (hashbitlen / 8) ) return BAD_HASHBITLEN;
|
||||
__m256i* x = (__m256i*)sp->x;
|
||||
|
||||
/* Sanity checks */
|
||||
if ( rounds <= 0 || rounds > 32 )
|
||||
rounds = CUBEHASH_ROUNDS;
|
||||
if ( blockbytes <= 0 || blockbytes >= 256)
|
||||
blockbytes = CUBEHASH_BLOCKBYTES;
|
||||
x[0] = _mm256_set_epi64x( iv[ 3], iv[ 2], iv[ 1], iv[ 0] );
|
||||
x[1] = _mm256_set_epi64x( iv[ 7], iv[ 6], iv[ 5], iv[ 4] );
|
||||
x[2] = _mm256_set_epi64x( iv[11], iv[10], iv[ 9], iv[ 8] );
|
||||
x[3] = _mm256_set_epi64x( iv[15], iv[14], iv[13], iv[12] );
|
||||
|
||||
// all sizes of __m128i
|
||||
cube_ctx_cache.hashlen = hashbitlen/128;
|
||||
cube_ctx_cache.blocksize = blockbytes/16;
|
||||
cube_ctx_cache.rounds = rounds;
|
||||
cube_ctx_cache.pos = 0;
|
||||
#else
|
||||
|
||||
for ( i = 0; i < 8; ++i )
|
||||
cube_ctx_cache.x[i] = _mm_setzero_si128();;
|
||||
__m128i* x = (__m128i*)sp->x;
|
||||
|
||||
cube_ctx_cache.x[0] = _mm_set_epi32( 0, rounds, blockbytes,
|
||||
hashbitlen / 8 );
|
||||
x[0] = _mm_set_epi64x( iv[ 1], iv[ 0] );
|
||||
x[1] = _mm_set_epi64x( iv[ 3], iv[ 2] );
|
||||
x[2] = _mm_set_epi64x( iv[ 5], iv[ 4] );
|
||||
x[3] = _mm_set_epi64x( iv[ 7], iv[ 6] );
|
||||
x[4] = _mm_set_epi64x( iv[ 9], iv[ 8] );
|
||||
x[5] = _mm_set_epi64x( iv[11], iv[10] );
|
||||
x[6] = _mm_set_epi64x( iv[13], iv[12] );
|
||||
x[7] = _mm_set_epi64x( iv[15], iv[14] );
|
||||
|
||||
for ( i = 0; i < 10; ++i )
|
||||
transform( &cube_ctx_cache );
|
||||
|
||||
memcpy( sp, &cube_ctx_cache, sizeof(cubehashParam) );
|
||||
#endif
|
||||
return SUCCESS;
|
||||
}
|
||||
|
||||
@@ -255,6 +262,7 @@ int cubehashUpdateDigest( cubehashParam *sp, byte *digest,
|
||||
transform( sp );
|
||||
|
||||
sp->x[7] = _mm_xor_si128( sp->x[7], _mm_set_epi32( 1,0,0,0 ) );
|
||||
|
||||
transform( sp );
|
||||
transform( sp );
|
||||
transform( sp );
|
||||
@@ -60,336 +60,174 @@ MYALIGN const unsigned int zero[] = {0x00000000, 0x00000000, 0x00000000, 0x000
|
||||
MYALIGN const unsigned int mul2ipt[] = {0x728efc00, 0x6894e61a, 0x3fc3b14d, 0x25d9ab57, 0xfd5ba600, 0x2a8c71d7, 0x1eb845e3, 0xc96f9234};
|
||||
|
||||
|
||||
//#include "crypto_hash.h"
|
||||
|
||||
int crypto_hash(
|
||||
unsigned char *out,
|
||||
const unsigned char *in,
|
||||
unsigned long long inlen
|
||||
)
|
||||
{
|
||||
|
||||
if(hash_echo(512, in, inlen * 8, out) == SUCCESS)
|
||||
return 0;
|
||||
|
||||
return -1;
|
||||
}
|
||||
|
||||
/*
|
||||
int main()
|
||||
{
|
||||
return 0;
|
||||
}
|
||||
*/
|
||||
|
||||
#if 0
|
||||
void DumpState(__m128i *ps)
|
||||
{
|
||||
int i, j, k;
|
||||
unsigned int ucol;
|
||||
|
||||
for(j = 0; j < 4; j++)
|
||||
{
|
||||
for(i = 0; i < 4; i++)
|
||||
{
|
||||
printf("row %d,col %d : ", i, j);
|
||||
for(k = 0; k < 4; k++)
|
||||
{
|
||||
ucol = *((int*)ps + 16 * i + 4 * j + k);
|
||||
printf("%02x%02x%02x%02x ", (ucol >> 0) & 0xff, (ucol >> 8) & 0xff, (ucol >> 16) & 0xff, (ucol >> 24) & 0xff);
|
||||
}
|
||||
|
||||
printf("\n");
|
||||
}
|
||||
}
|
||||
|
||||
printf("\n");
|
||||
}
|
||||
#endif
|
||||
|
||||
|
||||
|
||||
|
||||
#ifndef NO_AES_NI
|
||||
#define ECHO_SUBBYTES(state, i, j) \
|
||||
state[i][j] = _mm_aesenc_si128(state[i][j], k1);\
|
||||
state[i][j] = _mm_aesenc_si128(state[i][j], M128(zero));\
|
||||
k1 = _mm_add_epi32(k1, M128(const1))
|
||||
#else
|
||||
#define ECHO_SUBBYTES(state, i, j) \
|
||||
AES_ROUND_VPERM(state[i][j], t1, t2, t3, t4, s1, s2, s3);\
|
||||
state[i][j] = _mm_xor_si128(state[i][j], k1);\
|
||||
AES_ROUND_VPERM(state[i][j], t1, t2, t3, t4, s1, s2, s3);\
|
||||
k1 = _mm_add_epi32(k1, M128(const1))
|
||||
|
||||
#define ECHO_SUB_AND_MIX(state, i, j, state2, c, r1, r2, r3, r4) \
|
||||
AES_ROUND_VPERM_CORE(state[i][j], t1, t2, t3, t4, s1, s2, s3);\
|
||||
ktemp = k1;\
|
||||
TRANSFORM(ktemp, _k_ipt, t1, t4);\
|
||||
state[i][j] = _mm_xor_si128(state[i][j], ktemp);\
|
||||
AES_ROUND_VPERM_CORE(state[i][j], t1, t2, t3, t4, s1, s2, s3);\
|
||||
k1 = _mm_add_epi32(k1, M128(const1));\
|
||||
s1 = state[i][j];\
|
||||
s2 = s1;\
|
||||
TRANSFORM(s2, mul2ipt, t1, t2);\
|
||||
s3 = _mm_xor_si128(s1, s2);\
|
||||
state2[r1][c] = _mm_xor_si128(state2[r1][c], s2);\
|
||||
state2[r2][c] = _mm_xor_si128(state2[r2][c], s1);\
|
||||
state2[r3][c] = _mm_xor_si128(state2[r3][c], s1);\
|
||||
state2[r4][c] = _mm_xor_si128(state2[r4][c], s3)
|
||||
|
||||
|
||||
|
||||
#endif
|
||||
|
||||
state[i][j] = _mm_aesenc_si128(state[i][j], k1);\
|
||||
state[i][j] = _mm_aesenc_si128(state[i][j], M128(zero));\
|
||||
k1 = _mm_add_epi32(k1, M128(const1))
|
||||
|
||||
#define ECHO_MIXBYTES(state1, state2, j, t1, t2, s2) \
|
||||
s2 = _mm_add_epi8(state1[0][j], state1[0][j]);\
|
||||
t1 = _mm_srli_epi16(state1[0][j], 7);\
|
||||
t1 = _mm_and_si128(t1, M128(lsbmask));\
|
||||
t2 = _mm_shuffle_epi8(M128(mul2mask), t1);\
|
||||
s2 = _mm_xor_si128(s2, t2);\
|
||||
state2[0][j] = s2;\
|
||||
state2[1][j] = state1[0][j];\
|
||||
state2[2][j] = state1[0][j];\
|
||||
state2[3][j] = _mm_xor_si128(s2, state1[0][j]);\
|
||||
s2 = _mm_add_epi8(state1[1][(j + 1) & 3], state1[1][(j + 1) & 3]);\
|
||||
t1 = _mm_srli_epi16(state1[1][(j + 1) & 3], 7);\
|
||||
t1 = _mm_and_si128(t1, M128(lsbmask));\
|
||||
t2 = _mm_shuffle_epi8(M128(mul2mask), t1);\
|
||||
s2 = _mm_xor_si128(s2, t2);\
|
||||
state2[0][j] = _mm_xor_si128(state2[0][j], _mm_xor_si128(s2, state1[1][(j + 1) & 3]));\
|
||||
state2[1][j] = _mm_xor_si128(state2[1][j], s2);\
|
||||
state2[2][j] = _mm_xor_si128(state2[2][j], state1[1][(j + 1) & 3]);\
|
||||
state2[3][j] = _mm_xor_si128(state2[3][j], state1[1][(j + 1) & 3]);\
|
||||
s2 = _mm_add_epi8(state1[2][(j + 2) & 3], state1[2][(j + 2) & 3]);\
|
||||
t1 = _mm_srli_epi16(state1[2][(j + 2) & 3], 7);\
|
||||
t1 = _mm_and_si128(t1, M128(lsbmask));\
|
||||
t2 = _mm_shuffle_epi8(M128(mul2mask), t1);\
|
||||
s2 = _mm_xor_si128(s2, t2);\
|
||||
state2[0][j] = _mm_xor_si128(state2[0][j], state1[2][(j + 2) & 3]);\
|
||||
state2[1][j] = _mm_xor_si128(state2[1][j], _mm_xor_si128(s2, state1[2][(j + 2) & 3]));\
|
||||
state2[2][j] = _mm_xor_si128(state2[2][j], s2);\
|
||||
state2[3][j] = _mm_xor_si128(state2[3][j], state1[2][(j + 2) & 3]);\
|
||||
s2 = _mm_add_epi8(state1[3][(j + 3) & 3], state1[3][(j + 3) & 3]);\
|
||||
t1 = _mm_srli_epi16(state1[3][(j + 3) & 3], 7);\
|
||||
t1 = _mm_and_si128(t1, M128(lsbmask));\
|
||||
t2 = _mm_shuffle_epi8(M128(mul2mask), t1);\
|
||||
s2 = _mm_xor_si128(s2, t2);\
|
||||
state2[0][j] = _mm_xor_si128(state2[0][j], state1[3][(j + 3) & 3]);\
|
||||
state2[1][j] = _mm_xor_si128(state2[1][j], state1[3][(j + 3) & 3]);\
|
||||
state2[2][j] = _mm_xor_si128(state2[2][j], _mm_xor_si128(s2, state1[3][(j + 3) & 3]));\
|
||||
state2[3][j] = _mm_xor_si128(state2[3][j], s2)
|
||||
s2 = _mm_add_epi8(state1[0][j], state1[0][j]);\
|
||||
t1 = _mm_srli_epi16(state1[0][j], 7);\
|
||||
t1 = _mm_and_si128(t1, M128(lsbmask));\
|
||||
t2 = _mm_shuffle_epi8(M128(mul2mask), t1);\
|
||||
s2 = _mm_xor_si128(s2, t2);\
|
||||
state2[0][j] = s2;\
|
||||
state2[1][j] = state1[0][j];\
|
||||
state2[2][j] = state1[0][j];\
|
||||
state2[3][j] = _mm_xor_si128(s2, state1[0][j]);\
|
||||
s2 = _mm_add_epi8(state1[1][(j + 1) & 3], state1[1][(j + 1) & 3]);\
|
||||
t1 = _mm_srli_epi16(state1[1][(j + 1) & 3], 7);\
|
||||
t1 = _mm_and_si128(t1, M128(lsbmask));\
|
||||
t2 = _mm_shuffle_epi8(M128(mul2mask), t1);\
|
||||
s2 = _mm_xor_si128(s2, t2);\
|
||||
state2[0][j] = _mm_xor_si128(state2[0][j], _mm_xor_si128(s2, state1[1][(j + 1) & 3]));\
|
||||
state2[1][j] = _mm_xor_si128(state2[1][j], s2);\
|
||||
state2[2][j] = _mm_xor_si128(state2[2][j], state1[1][(j + 1) & 3]);\
|
||||
state2[3][j] = _mm_xor_si128(state2[3][j], state1[1][(j + 1) & 3]);\
|
||||
s2 = _mm_add_epi8(state1[2][(j + 2) & 3], state1[2][(j + 2) & 3]);\
|
||||
t1 = _mm_srli_epi16(state1[2][(j + 2) & 3], 7);\
|
||||
t1 = _mm_and_si128(t1, M128(lsbmask));\
|
||||
t2 = _mm_shuffle_epi8(M128(mul2mask), t1);\
|
||||
s2 = _mm_xor_si128(s2, t2);\
|
||||
state2[0][j] = _mm_xor_si128(state2[0][j], state1[2][(j + 2) & 3]);\
|
||||
state2[1][j] = _mm_xor_si128(state2[1][j], _mm_xor_si128(s2, state1[2][(j + 2) & 3]));\
|
||||
state2[2][j] = _mm_xor_si128(state2[2][j], s2);\
|
||||
state2[3][j] = _mm_xor_si128(state2[3][j], state1[2][(j + 2) & 3]);\
|
||||
s2 = _mm_add_epi8(state1[3][(j + 3) & 3], state1[3][(j + 3) & 3]);\
|
||||
t1 = _mm_srli_epi16(state1[3][(j + 3) & 3], 7);\
|
||||
t1 = _mm_and_si128(t1, M128(lsbmask));\
|
||||
t2 = _mm_shuffle_epi8(M128(mul2mask), t1);\
|
||||
s2 = _mm_xor_si128(s2, t2);\
|
||||
state2[0][j] = _mm_xor_si128(state2[0][j], state1[3][(j + 3) & 3]);\
|
||||
state2[1][j] = _mm_xor_si128(state2[1][j], state1[3][(j + 3) & 3]);\
|
||||
state2[2][j] = _mm_xor_si128(state2[2][j], _mm_xor_si128(s2, state1[3][(j + 3) & 3]));\
|
||||
state2[3][j] = _mm_xor_si128(state2[3][j], s2)
|
||||
|
||||
|
||||
#define ECHO_ROUND_UNROLL2 \
|
||||
ECHO_SUBBYTES(_state, 0, 0);\
|
||||
ECHO_SUBBYTES(_state, 1, 0);\
|
||||
ECHO_SUBBYTES(_state, 2, 0);\
|
||||
ECHO_SUBBYTES(_state, 3, 0);\
|
||||
ECHO_SUBBYTES(_state, 0, 1);\
|
||||
ECHO_SUBBYTES(_state, 1, 1);\
|
||||
ECHO_SUBBYTES(_state, 2, 1);\
|
||||
ECHO_SUBBYTES(_state, 3, 1);\
|
||||
ECHO_SUBBYTES(_state, 0, 2);\
|
||||
ECHO_SUBBYTES(_state, 1, 2);\
|
||||
ECHO_SUBBYTES(_state, 2, 2);\
|
||||
ECHO_SUBBYTES(_state, 3, 2);\
|
||||
ECHO_SUBBYTES(_state, 0, 3);\
|
||||
ECHO_SUBBYTES(_state, 1, 3);\
|
||||
ECHO_SUBBYTES(_state, 2, 3);\
|
||||
ECHO_SUBBYTES(_state, 3, 3);\
|
||||
ECHO_MIXBYTES(_state, _state2, 0, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state, _state2, 1, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state, _state2, 2, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state, _state2, 3, t1, t2, s2);\
|
||||
ECHO_SUBBYTES(_state2, 0, 0);\
|
||||
ECHO_SUBBYTES(_state2, 1, 0);\
|
||||
ECHO_SUBBYTES(_state2, 2, 0);\
|
||||
ECHO_SUBBYTES(_state2, 3, 0);\
|
||||
ECHO_SUBBYTES(_state2, 0, 1);\
|
||||
ECHO_SUBBYTES(_state2, 1, 1);\
|
||||
ECHO_SUBBYTES(_state2, 2, 1);\
|
||||
ECHO_SUBBYTES(_state2, 3, 1);\
|
||||
ECHO_SUBBYTES(_state2, 0, 2);\
|
||||
ECHO_SUBBYTES(_state2, 1, 2);\
|
||||
ECHO_SUBBYTES(_state2, 2, 2);\
|
||||
ECHO_SUBBYTES(_state2, 3, 2);\
|
||||
ECHO_SUBBYTES(_state2, 0, 3);\
|
||||
ECHO_SUBBYTES(_state2, 1, 3);\
|
||||
ECHO_SUBBYTES(_state2, 2, 3);\
|
||||
ECHO_SUBBYTES(_state2, 3, 3);\
|
||||
ECHO_MIXBYTES(_state2, _state, 0, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state2, _state, 1, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state2, _state, 2, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state2, _state, 3, t1, t2, s2)
|
||||
ECHO_SUBBYTES(_state, 0, 0);\
|
||||
ECHO_SUBBYTES(_state, 1, 0);\
|
||||
ECHO_SUBBYTES(_state, 2, 0);\
|
||||
ECHO_SUBBYTES(_state, 3, 0);\
|
||||
ECHO_SUBBYTES(_state, 0, 1);\
|
||||
ECHO_SUBBYTES(_state, 1, 1);\
|
||||
ECHO_SUBBYTES(_state, 2, 1);\
|
||||
ECHO_SUBBYTES(_state, 3, 1);\
|
||||
ECHO_SUBBYTES(_state, 0, 2);\
|
||||
ECHO_SUBBYTES(_state, 1, 2);\
|
||||
ECHO_SUBBYTES(_state, 2, 2);\
|
||||
ECHO_SUBBYTES(_state, 3, 2);\
|
||||
ECHO_SUBBYTES(_state, 0, 3);\
|
||||
ECHO_SUBBYTES(_state, 1, 3);\
|
||||
ECHO_SUBBYTES(_state, 2, 3);\
|
||||
ECHO_SUBBYTES(_state, 3, 3);\
|
||||
ECHO_MIXBYTES(_state, _state2, 0, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state, _state2, 1, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state, _state2, 2, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state, _state2, 3, t1, t2, s2);\
|
||||
ECHO_SUBBYTES(_state2, 0, 0);\
|
||||
ECHO_SUBBYTES(_state2, 1, 0);\
|
||||
ECHO_SUBBYTES(_state2, 2, 0);\
|
||||
ECHO_SUBBYTES(_state2, 3, 0);\
|
||||
ECHO_SUBBYTES(_state2, 0, 1);\
|
||||
ECHO_SUBBYTES(_state2, 1, 1);\
|
||||
ECHO_SUBBYTES(_state2, 2, 1);\
|
||||
ECHO_SUBBYTES(_state2, 3, 1);\
|
||||
ECHO_SUBBYTES(_state2, 0, 2);\
|
||||
ECHO_SUBBYTES(_state2, 1, 2);\
|
||||
ECHO_SUBBYTES(_state2, 2, 2);\
|
||||
ECHO_SUBBYTES(_state2, 3, 2);\
|
||||
ECHO_SUBBYTES(_state2, 0, 3);\
|
||||
ECHO_SUBBYTES(_state2, 1, 3);\
|
||||
ECHO_SUBBYTES(_state2, 2, 3);\
|
||||
ECHO_SUBBYTES(_state2, 3, 3);\
|
||||
ECHO_MIXBYTES(_state2, _state, 0, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state2, _state, 1, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state2, _state, 2, t1, t2, s2);\
|
||||
ECHO_MIXBYTES(_state2, _state, 3, t1, t2, s2)
|
||||
|
||||
|
||||
|
||||
#define SAVESTATE(dst, src)\
|
||||
dst[0][0] = src[0][0];\
|
||||
dst[0][1] = src[0][1];\
|
||||
dst[0][2] = src[0][2];\
|
||||
dst[0][3] = src[0][3];\
|
||||
dst[1][0] = src[1][0];\
|
||||
dst[1][1] = src[1][1];\
|
||||
dst[1][2] = src[1][2];\
|
||||
dst[1][3] = src[1][3];\
|
||||
dst[2][0] = src[2][0];\
|
||||
dst[2][1] = src[2][1];\
|
||||
dst[2][2] = src[2][2];\
|
||||
dst[2][3] = src[2][3];\
|
||||
dst[3][0] = src[3][0];\
|
||||
dst[3][1] = src[3][1];\
|
||||
dst[3][2] = src[3][2];\
|
||||
dst[3][3] = src[3][3]
|
||||
dst[0][0] = src[0][0];\
|
||||
dst[0][1] = src[0][1];\
|
||||
dst[0][2] = src[0][2];\
|
||||
dst[0][3] = src[0][3];\
|
||||
dst[1][0] = src[1][0];\
|
||||
dst[1][1] = src[1][1];\
|
||||
dst[1][2] = src[1][2];\
|
||||
dst[1][3] = src[1][3];\
|
||||
dst[2][0] = src[2][0];\
|
||||
dst[2][1] = src[2][1];\
|
||||
dst[2][2] = src[2][2];\
|
||||
dst[2][3] = src[2][3];\
|
||||
dst[3][0] = src[3][0];\
|
||||
dst[3][1] = src[3][1];\
|
||||
dst[3][2] = src[3][2];\
|
||||
dst[3][3] = src[3][3]
|
||||
|
||||
|
||||
void Compress(hashState_echo *ctx, const unsigned char *pmsg, unsigned int uBlockCount)
|
||||
{
|
||||
unsigned int r, b, i, j;
|
||||
// __m128i t1, t2, t3, t4, s1, s2, s3, k1, ktemp;
|
||||
__m128i t1, t2, s2, k1;
|
||||
__m128i _state[4][4], _state2[4][4], _statebackup[4][4];
|
||||
unsigned int r, b, i, j;
|
||||
__m128i t1, t2, s2, k1;
|
||||
__m128i _state[4][4], _state2[4][4], _statebackup[4][4];
|
||||
|
||||
for(i = 0; i < 4; i++)
|
||||
for(j = 0; j < ctx->uHashSize / 256; j++)
|
||||
_state[i][j] = ctx->state[i][j];
|
||||
|
||||
for(i = 0; i < 4; i++)
|
||||
for(j = 0; j < ctx->uHashSize / 256; j++)
|
||||
_state[i][j] = ctx->state[i][j];
|
||||
for(b = 0; b < uBlockCount; b++)
|
||||
{
|
||||
ctx->k = _mm_add_epi64(ctx->k, ctx->const1536);
|
||||
|
||||
|
||||
#ifdef NO_AES_NI
|
||||
// transform cv
|
||||
for(i = 0; i < 4; i++)
|
||||
for(j = 0; j < ctx->uHashSize / 256; j++)
|
||||
{
|
||||
TRANSFORM(_state[i][j], _k_ipt, t1, t2);
|
||||
}
|
||||
#endif
|
||||
|
||||
for(b = 0; b < uBlockCount; b++)
|
||||
// load message
|
||||
for(j = ctx->uHashSize / 256; j < 4; j++)
|
||||
{
|
||||
ctx->k = _mm_add_epi64(ctx->k, ctx->const1536);
|
||||
|
||||
// load message
|
||||
for(j = ctx->uHashSize / 256; j < 4; j++)
|
||||
{
|
||||
for(i = 0; i < 4; i++)
|
||||
{
|
||||
_state[i][j] = _mm_loadu_si128((__m128i*)pmsg + 4 * (j - (ctx->uHashSize / 256)) + i);
|
||||
|
||||
#ifdef NO_AES_NI
|
||||
// transform message
|
||||
TRANSFORM(_state[i][j], _k_ipt, t1, t2);
|
||||
#endif
|
||||
}
|
||||
}
|
||||
|
||||
// save state
|
||||
SAVESTATE(_statebackup, _state);
|
||||
|
||||
|
||||
k1 = ctx->k;
|
||||
|
||||
#ifndef NO_AES_NI
|
||||
for(r = 0; r < ctx->uRounds / 2; r++)
|
||||
{
|
||||
ECHO_ROUND_UNROLL2;
|
||||
}
|
||||
|
||||
#else
|
||||
for(r = 0; r < ctx->uRounds / 2; r++)
|
||||
{
|
||||
_state2[0][0] = M128(zero); _state2[1][0] = M128(zero); _state2[2][0] = M128(zero); _state2[3][0] = M128(zero);
|
||||
_state2[0][1] = M128(zero); _state2[1][1] = M128(zero); _state2[2][1] = M128(zero); _state2[3][1] = M128(zero);
|
||||
_state2[0][2] = M128(zero); _state2[1][2] = M128(zero); _state2[2][2] = M128(zero); _state2[3][2] = M128(zero);
|
||||
_state2[0][3] = M128(zero); _state2[1][3] = M128(zero); _state2[2][3] = M128(zero); _state2[3][3] = M128(zero);
|
||||
|
||||
ECHO_SUB_AND_MIX(_state, 0, 0, _state2, 0, 0, 1, 2, 3);
|
||||
ECHO_SUB_AND_MIX(_state, 1, 0, _state2, 3, 1, 2, 3, 0);
|
||||
ECHO_SUB_AND_MIX(_state, 2, 0, _state2, 2, 2, 3, 0, 1);
|
||||
ECHO_SUB_AND_MIX(_state, 3, 0, _state2, 1, 3, 0, 1, 2);
|
||||
ECHO_SUB_AND_MIX(_state, 0, 1, _state2, 1, 0, 1, 2, 3);
|
||||
ECHO_SUB_AND_MIX(_state, 1, 1, _state2, 0, 1, 2, 3, 0);
|
||||
ECHO_SUB_AND_MIX(_state, 2, 1, _state2, 3, 2, 3, 0, 1);
|
||||
ECHO_SUB_AND_MIX(_state, 3, 1, _state2, 2, 3, 0, 1, 2);
|
||||
ECHO_SUB_AND_MIX(_state, 0, 2, _state2, 2, 0, 1, 2, 3);
|
||||
ECHO_SUB_AND_MIX(_state, 1, 2, _state2, 1, 1, 2, 3, 0);
|
||||
ECHO_SUB_AND_MIX(_state, 2, 2, _state2, 0, 2, 3, 0, 1);
|
||||
ECHO_SUB_AND_MIX(_state, 3, 2, _state2, 3, 3, 0, 1, 2);
|
||||
ECHO_SUB_AND_MIX(_state, 0, 3, _state2, 3, 0, 1, 2, 3);
|
||||
ECHO_SUB_AND_MIX(_state, 1, 3, _state2, 2, 1, 2, 3, 0);
|
||||
ECHO_SUB_AND_MIX(_state, 2, 3, _state2, 1, 2, 3, 0, 1);
|
||||
ECHO_SUB_AND_MIX(_state, 3, 3, _state2, 0, 3, 0, 1, 2);
|
||||
|
||||
_state[0][0] = M128(zero); _state[1][0] = M128(zero); _state[2][0] = M128(zero); _state[3][0] = M128(zero);
|
||||
_state[0][1] = M128(zero); _state[1][1] = M128(zero); _state[2][1] = M128(zero); _state[3][1] = M128(zero);
|
||||
_state[0][2] = M128(zero); _state[1][2] = M128(zero); _state[2][2] = M128(zero); _state[3][2] = M128(zero);
|
||||
_state[0][3] = M128(zero); _state[1][3] = M128(zero); _state[2][3] = M128(zero); _state[3][3] = M128(zero);
|
||||
|
||||
ECHO_SUB_AND_MIX(_state2, 0, 0, _state, 0, 0, 1, 2, 3);
|
||||
ECHO_SUB_AND_MIX(_state2, 1, 0, _state, 3, 1, 2, 3, 0);
|
||||
ECHO_SUB_AND_MIX(_state2, 2, 0, _state, 2, 2, 3, 0, 1);
|
||||
ECHO_SUB_AND_MIX(_state2, 3, 0, _state, 1, 3, 0, 1, 2);
|
||||
ECHO_SUB_AND_MIX(_state2, 0, 1, _state, 1, 0, 1, 2, 3);
|
||||
ECHO_SUB_AND_MIX(_state2, 1, 1, _state, 0, 1, 2, 3, 0);
|
||||
ECHO_SUB_AND_MIX(_state2, 2, 1, _state, 3, 2, 3, 0, 1);
|
||||
ECHO_SUB_AND_MIX(_state2, 3, 1, _state, 2, 3, 0, 1, 2);
|
||||
ECHO_SUB_AND_MIX(_state2, 0, 2, _state, 2, 0, 1, 2, 3);
|
||||
ECHO_SUB_AND_MIX(_state2, 1, 2, _state, 1, 1, 2, 3, 0);
|
||||
ECHO_SUB_AND_MIX(_state2, 2, 2, _state, 0, 2, 3, 0, 1);
|
||||
ECHO_SUB_AND_MIX(_state2, 3, 2, _state, 3, 3, 0, 1, 2);
|
||||
ECHO_SUB_AND_MIX(_state2, 0, 3, _state, 3, 0, 1, 2, 3);
|
||||
ECHO_SUB_AND_MIX(_state2, 1, 3, _state, 2, 1, 2, 3, 0);
|
||||
ECHO_SUB_AND_MIX(_state2, 2, 3, _state, 1, 2, 3, 0, 1);
|
||||
ECHO_SUB_AND_MIX(_state2, 3, 3, _state, 0, 3, 0, 1, 2);
|
||||
|
||||
}
|
||||
#endif
|
||||
|
||||
|
||||
if(ctx->uHashSize == 256)
|
||||
{
|
||||
for(i = 0; i < 4; i++)
|
||||
{
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _state[i][1]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _state[i][2]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _state[i][3]);
|
||||
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][0]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][1]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][2]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][3]);
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
for(i = 0; i < 4; i++)
|
||||
{
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _state[i][2]);
|
||||
_state[i][1] = _mm_xor_si128(_state[i][1], _state[i][3]);
|
||||
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][0]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][2]);
|
||||
|
||||
_state[i][1] = _mm_xor_si128(_state[i][1], _statebackup[i][1]);
|
||||
_state[i][1] = _mm_xor_si128(_state[i][1], _statebackup[i][3]);
|
||||
}
|
||||
}
|
||||
|
||||
pmsg += ctx->uBlockLength;
|
||||
for(i = 0; i < 4; i++)
|
||||
{
|
||||
_state[i][j] = _mm_loadu_si128((__m128i*)pmsg + 4 * (j - (ctx->uHashSize / 256)) + i);
|
||||
}
|
||||
}
|
||||
|
||||
#ifdef NO_AES_NI
|
||||
// transform state
|
||||
for(i = 0; i < 4; i++)
|
||||
for(j = 0; j < 4; j++)
|
||||
{
|
||||
TRANSFORM(_state[i][j], _k_opt, t1, t2);
|
||||
}
|
||||
#endif
|
||||
// save state
|
||||
SAVESTATE(_statebackup, _state);
|
||||
|
||||
SAVESTATE(ctx->state, _state);
|
||||
k1 = ctx->k;
|
||||
|
||||
for(r = 0; r < ctx->uRounds / 2; r++)
|
||||
{
|
||||
ECHO_ROUND_UNROLL2;
|
||||
}
|
||||
|
||||
if(ctx->uHashSize == 256)
|
||||
{
|
||||
for(i = 0; i < 4; i++)
|
||||
{
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _state[i][1]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _state[i][2]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _state[i][3]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][0]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][1]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][2]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][3]);
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
for(i = 0; i < 4; i++)
|
||||
{
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _state[i][2]);
|
||||
_state[i][1] = _mm_xor_si128(_state[i][1], _state[i][3]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][0]);
|
||||
_state[i][0] = _mm_xor_si128(_state[i][0], _statebackup[i][2]);
|
||||
_state[i][1] = _mm_xor_si128(_state[i][1], _statebackup[i][1]);
|
||||
_state[i][1] = _mm_xor_si128(_state[i][1], _statebackup[i][3]);
|
||||
}
|
||||
}
|
||||
pmsg += ctx->uBlockLength;
|
||||
}
|
||||
SAVESTATE(ctx->state, _state);
|
||||
|
||||
}
|
||||
|
||||
|
||||
@@ -30,6 +30,7 @@
|
||||
typedef struct
|
||||
{
|
||||
__m128i state[4][4];
|
||||
BitSequence buffer[192];
|
||||
__m128i k;
|
||||
__m128i hashsize;
|
||||
__m128i const1536;
|
||||
@@ -39,9 +40,8 @@ typedef struct
|
||||
unsigned int uBlockLength;
|
||||
unsigned int uBufferBytes;
|
||||
DataLength processed_bits;
|
||||
BitSequence buffer[192];
|
||||
|
||||
} hashState_echo;
|
||||
} hashState_echo __attribute__ ((aligned (64)));
|
||||
|
||||
HashReturn init_echo(hashState_echo *state, int hashbitlen);
|
||||
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@@ -1,320 +0,0 @@
|
||||
/* $Id: sph_echo.h 216 2010-06-08 09:46:57Z tp $ */
|
||||
/**
|
||||
* ECHO interface. ECHO is a family of functions which differ by
|
||||
* their output size; this implementation defines ECHO for output
|
||||
* sizes 224, 256, 384 and 512 bits.
|
||||
*
|
||||
* ==========================(LICENSE BEGIN)============================
|
||||
*
|
||||
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
|
||||
*
|
||||
* Permission is hereby granted, free of charge, to any person obtaining
|
||||
* a copy of this software and associated documentation files (the
|
||||
* "Software"), to deal in the Software without restriction, including
|
||||
* without limitation the rights to use, copy, modify, merge, publish,
|
||||
* distribute, sublicense, and/or sell copies of the Software, and to
|
||||
* permit persons to whom the Software is furnished to do so, subject to
|
||||
* the following conditions:
|
||||
*
|
||||
* The above copyright notice and this permission notice shall be
|
||||
* included in all copies or substantial portions of the Software.
|
||||
*
|
||||
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
||||
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
||||
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
|
||||
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
|
||||
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
|
||||
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
|
||||
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
||||
*
|
||||
* ===========================(LICENSE END)=============================
|
||||
*
|
||||
* @file sph_echo.h
|
||||
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
|
||||
*/
|
||||
|
||||
#ifndef SPH_ECHO_H__
|
||||
#define SPH_ECHO_H__
|
||||
|
||||
#ifdef __cplusplus
|
||||
extern "C"{
|
||||
#endif
|
||||
|
||||
#include <stddef.h>
|
||||
#include "algo/sha/sph_types.h"
|
||||
|
||||
/**
|
||||
* Output size (in bits) for ECHO-224.
|
||||
*/
|
||||
#define SPH_SIZE_echo224 224
|
||||
|
||||
/**
|
||||
* Output size (in bits) for ECHO-256.
|
||||
*/
|
||||
#define SPH_SIZE_echo256 256
|
||||
|
||||
/**
|
||||
* Output size (in bits) for ECHO-384.
|
||||
*/
|
||||
#define SPH_SIZE_echo384 384
|
||||
|
||||
/**
|
||||
* Output size (in bits) for ECHO-512.
|
||||
*/
|
||||
#define SPH_SIZE_echo512 512
|
||||
|
||||
/**
|
||||
* This structure is a context for ECHO computations: it contains the
|
||||
* intermediate values and some data from the last entered block. Once
|
||||
* an ECHO computation has been performed, the context can be reused for
|
||||
* another computation. This specific structure is used for ECHO-224
|
||||
* and ECHO-256.
|
||||
*
|
||||
* The contents of this structure are private. A running ECHO computation
|
||||
* can be cloned by copying the context (e.g. with a simple
|
||||
* <code>memcpy()</code>).
|
||||
*/
|
||||
typedef struct {
|
||||
#ifndef DOXYGEN_IGNORE
|
||||
unsigned char buf[192]; /* first field, for alignment */
|
||||
size_t ptr;
|
||||
union {
|
||||
sph_u32 Vs[4][4];
|
||||
#if SPH_64
|
||||
sph_u64 Vb[4][2];
|
||||
#endif
|
||||
} u;
|
||||
sph_u32 C0, C1, C2, C3;
|
||||
#endif
|
||||
} sph_echo_small_context;
|
||||
|
||||
/**
|
||||
* This structure is a context for ECHO computations: it contains the
|
||||
* intermediate values and some data from the last entered block. Once
|
||||
* an ECHO computation has been performed, the context can be reused for
|
||||
* another computation. This specific structure is used for ECHO-384
|
||||
* and ECHO-512.
|
||||
*
|
||||
* The contents of this structure are private. A running ECHO computation
|
||||
* can be cloned by copying the context (e.g. with a simple
|
||||
* <code>memcpy()</code>).
|
||||
*/
|
||||
typedef struct {
|
||||
#ifndef DOXYGEN_IGNORE
|
||||
unsigned char buf[128]; /* first field, for alignment */
|
||||
size_t ptr;
|
||||
union {
|
||||
sph_u32 Vs[8][4];
|
||||
#if SPH_64
|
||||
sph_u64 Vb[8][2];
|
||||
#endif
|
||||
} u;
|
||||
sph_u32 C0, C1, C2, C3;
|
||||
#endif
|
||||
} sph_echo_big_context;
|
||||
|
||||
/**
|
||||
* Type for a ECHO-224 context (identical to the common "small" context).
|
||||
*/
|
||||
typedef sph_echo_small_context sph_echo224_context;
|
||||
|
||||
/**
|
||||
* Type for a ECHO-256 context (identical to the common "small" context).
|
||||
*/
|
||||
typedef sph_echo_small_context sph_echo256_context;
|
||||
|
||||
/**
|
||||
* Type for a ECHO-384 context (identical to the common "big" context).
|
||||
*/
|
||||
typedef sph_echo_big_context sph_echo384_context;
|
||||
|
||||
/**
|
||||
* Type for a ECHO-512 context (identical to the common "big" context).
|
||||
*/
|
||||
typedef sph_echo_big_context sph_echo512_context;
|
||||
|
||||
/**
|
||||
* Initialize an ECHO-224 context. This process performs no memory allocation.
|
||||
*
|
||||
* @param cc the ECHO-224 context (pointer to a
|
||||
* <code>sph_echo224_context</code>)
|
||||
*/
|
||||
void sph_echo224_init(void *cc);
|
||||
|
||||
/**
|
||||
* Process some data bytes. It is acceptable that <code>len</code> is zero
|
||||
* (in which case this function does nothing).
|
||||
*
|
||||
* @param cc the ECHO-224 context
|
||||
* @param data the input data
|
||||
* @param len the input data length (in bytes)
|
||||
*/
|
||||
void sph_echo224(void *cc, const void *data, size_t len);
|
||||
|
||||
/**
|
||||
* Terminate the current ECHO-224 computation and output the result into
|
||||
* the provided buffer. The destination buffer must be wide enough to
|
||||
* accomodate the result (28 bytes). The context is automatically
|
||||
* reinitialized.
|
||||
*
|
||||
* @param cc the ECHO-224 context
|
||||
* @param dst the destination buffer
|
||||
*/
|
||||
void sph_echo224_close(void *cc, void *dst);
|
||||
|
||||
/**
|
||||
* Add a few additional bits (0 to 7) to the current computation, then
|
||||
* terminate it and output the result in the provided buffer, which must
|
||||
* be wide enough to accomodate the result (28 bytes). If bit number i
|
||||
* in <code>ub</code> has value 2^i, then the extra bits are those
|
||||
* numbered 7 downto 8-n (this is the big-endian convention at the byte
|
||||
* level). The context is automatically reinitialized.
|
||||
*
|
||||
* @param cc the ECHO-224 context
|
||||
* @param ub the extra bits
|
||||
* @param n the number of extra bits (0 to 7)
|
||||
* @param dst the destination buffer
|
||||
*/
|
||||
void sph_echo224_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
/**
|
||||
* Initialize an ECHO-256 context. This process performs no memory allocation.
|
||||
*
|
||||
* @param cc the ECHO-256 context (pointer to a
|
||||
* <code>sph_echo256_context</code>)
|
||||
*/
|
||||
void sph_echo256_init(void *cc);
|
||||
|
||||
/**
|
||||
* Process some data bytes. It is acceptable that <code>len</code> is zero
|
||||
* (in which case this function does nothing).
|
||||
*
|
||||
* @param cc the ECHO-256 context
|
||||
* @param data the input data
|
||||
* @param len the input data length (in bytes)
|
||||
*/
|
||||
void sph_echo256(void *cc, const void *data, size_t len);
|
||||
|
||||
/**
|
||||
* Terminate the current ECHO-256 computation and output the result into
|
||||
* the provided buffer. The destination buffer must be wide enough to
|
||||
* accomodate the result (32 bytes). The context is automatically
|
||||
* reinitialized.
|
||||
*
|
||||
* @param cc the ECHO-256 context
|
||||
* @param dst the destination buffer
|
||||
*/
|
||||
void sph_echo256_close(void *cc, void *dst);
|
||||
|
||||
/**
|
||||
* Add a few additional bits (0 to 7) to the current computation, then
|
||||
* terminate it and output the result in the provided buffer, which must
|
||||
* be wide enough to accomodate the result (32 bytes). If bit number i
|
||||
* in <code>ub</code> has value 2^i, then the extra bits are those
|
||||
* numbered 7 downto 8-n (this is the big-endian convention at the byte
|
||||
* level). The context is automatically reinitialized.
|
||||
*
|
||||
* @param cc the ECHO-256 context
|
||||
* @param ub the extra bits
|
||||
* @param n the number of extra bits (0 to 7)
|
||||
* @param dst the destination buffer
|
||||
*/
|
||||
void sph_echo256_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
/**
|
||||
* Initialize an ECHO-384 context. This process performs no memory allocation.
|
||||
*
|
||||
* @param cc the ECHO-384 context (pointer to a
|
||||
* <code>sph_echo384_context</code>)
|
||||
*/
|
||||
void sph_echo384_init(void *cc);
|
||||
|
||||
/**
|
||||
* Process some data bytes. It is acceptable that <code>len</code> is zero
|
||||
* (in which case this function does nothing).
|
||||
*
|
||||
* @param cc the ECHO-384 context
|
||||
* @param data the input data
|
||||
* @param len the input data length (in bytes)
|
||||
*/
|
||||
void sph_echo384(void *cc, const void *data, size_t len);
|
||||
|
||||
/**
|
||||
* Terminate the current ECHO-384 computation and output the result into
|
||||
* the provided buffer. The destination buffer must be wide enough to
|
||||
* accomodate the result (48 bytes). The context is automatically
|
||||
* reinitialized.
|
||||
*
|
||||
* @param cc the ECHO-384 context
|
||||
* @param dst the destination buffer
|
||||
*/
|
||||
void sph_echo384_close(void *cc, void *dst);
|
||||
|
||||
/**
|
||||
* Add a few additional bits (0 to 7) to the current computation, then
|
||||
* terminate it and output the result in the provided buffer, which must
|
||||
* be wide enough to accomodate the result (48 bytes). If bit number i
|
||||
* in <code>ub</code> has value 2^i, then the extra bits are those
|
||||
* numbered 7 downto 8-n (this is the big-endian convention at the byte
|
||||
* level). The context is automatically reinitialized.
|
||||
*
|
||||
* @param cc the ECHO-384 context
|
||||
* @param ub the extra bits
|
||||
* @param n the number of extra bits (0 to 7)
|
||||
* @param dst the destination buffer
|
||||
*/
|
||||
void sph_echo384_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
/**
|
||||
* Initialize an ECHO-512 context. This process performs no memory allocation.
|
||||
*
|
||||
* @param cc the ECHO-512 context (pointer to a
|
||||
* <code>sph_echo512_context</code>)
|
||||
*/
|
||||
void sph_echo512_init(void *cc);
|
||||
|
||||
/**
|
||||
* Process some data bytes. It is acceptable that <code>len</code> is zero
|
||||
* (in which case this function does nothing).
|
||||
*
|
||||
* @param cc the ECHO-512 context
|
||||
* @param data the input data
|
||||
* @param len the input data length (in bytes)
|
||||
*/
|
||||
void sph_echo512(void *cc, const void *data, size_t len);
|
||||
|
||||
/**
|
||||
* Terminate the current ECHO-512 computation and output the result into
|
||||
* the provided buffer. The destination buffer must be wide enough to
|
||||
* accomodate the result (64 bytes). The context is automatically
|
||||
* reinitialized.
|
||||
*
|
||||
* @param cc the ECHO-512 context
|
||||
* @param dst the destination buffer
|
||||
*/
|
||||
void sph_echo512_close(void *cc, void *dst);
|
||||
|
||||
/**
|
||||
* Add a few additional bits (0 to 7) to the current computation, then
|
||||
* terminate it and output the result in the provided buffer, which must
|
||||
* be wide enough to accomodate the result (64 bytes). If bit number i
|
||||
* in <code>ub</code> has value 2^i, then the extra bits are those
|
||||
* numbered 7 downto 8-n (this is the big-endian convention at the byte
|
||||
* level). The context is automatically reinitialized.
|
||||
*
|
||||
* @param cc the ECHO-512 context
|
||||
* @param ub the extra bits
|
||||
* @param n the number of extra bits (0 to 7)
|
||||
* @param dst the destination buffer
|
||||
*/
|
||||
void sph_echo512_addbits_and_close(
|
||||
void *cc, unsigned ub, unsigned n, void *dst);
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif
|
||||
@@ -11,6 +11,8 @@ extern "C"{
|
||||
#pragma warning (disable: 4146)
|
||||
#endif
|
||||
|
||||
#define SPH_FUGUE_NOCOPY 1
|
||||
|
||||
static const sph_u32 IV224[] = {
|
||||
SPH_C32(0xf4c9120d), SPH_C32(0x6286f757), SPH_C32(0xee39e01c),
|
||||
SPH_C32(0xe074e3cb), SPH_C32(0xa1127c62), SPH_C32(0x9a43d215),
|
||||
|
||||
@@ -12,7 +12,7 @@
|
||||
#include <memory.h>
|
||||
#include "hash-groestl.h"
|
||||
#include "miner.h"
|
||||
#include "avxdefs.h"
|
||||
#include "simd-utils.h"
|
||||
|
||||
#ifndef NO_AES_NI
|
||||
|
||||
|
||||
@@ -9,7 +9,7 @@
|
||||
#include <memory.h>
|
||||
#include "hash-groestl256.h"
|
||||
#include "miner.h"
|
||||
#include "avxdefs.h"
|
||||
#include "simd-utils.h"
|
||||
|
||||
#ifndef NO_AES_NI
|
||||
|
||||
|
||||
@@ -33,7 +33,7 @@ void myriad_4way_hash( void *output, const void *input )
|
||||
myrgr_4way_ctx_holder ctx;
|
||||
memcpy( &ctx, &myrgr_4way_ctx, sizeof(myrgr_4way_ctx) );
|
||||
|
||||
mm_deinterleave_4x32( hash0, hash1, hash2, hash3, input, 640 );
|
||||
mm128_deinterleave_4x32( hash0, hash1, hash2, hash3, input, 640 );
|
||||
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash0, (char*)hash0, 640 );
|
||||
memcpy( &ctx.groestl, &myrgr_4way_ctx.groestl, sizeof(hashState_groestl) );
|
||||
@@ -43,12 +43,12 @@ void myriad_4way_hash( void *output, const void *input )
|
||||
memcpy( &ctx.groestl, &myrgr_4way_ctx.groestl, sizeof(hashState_groestl) );
|
||||
update_and_final_groestl( &ctx.groestl, (char*)hash3, (char*)hash3, 640 );
|
||||
|
||||
mm_interleave_4x32( vhash, hash0, hash1, hash2, hash3, 512 );
|
||||
mm128_interleave_4x32( vhash, hash0, hash1, hash2, hash3, 512 );
|
||||
|
||||
sha256_4way( &ctx.sha, vhash, 64 );
|
||||
sha256_4way_close( &ctx.sha, vhash );
|
||||
|
||||
mm_deinterleave_4x32( output, output+32, output+64, output+96,
|
||||
mm128_deinterleave_4x32( output, output+32, output+64, output+96,
|
||||
vhash, 256 );
|
||||
}
|
||||
|
||||
@@ -79,7 +79,7 @@ int scanhash_myriad_4way( int thr_id, struct work *work, uint32_t max_nonce,
|
||||
( (uint32_t*)ptarget )[7] = 0x0000ff;
|
||||
|
||||
swab32_array( edata, pdata, 20 );
|
||||
mm_interleave_4x32( vdata, edata, edata, edata, edata, 640 );
|
||||
mm128_interleave_4x32( vdata, edata, edata, edata, edata, 640 );
|
||||
|
||||
do {
|
||||
be32enc( noncep, n );
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user